VII Dubrovnik Conference on Cognitive Science

DUBROVNIK, 21-23 MAY, 2015.

Metacognition and Reasoning

Invited Speakers:
Klaus Fiedler, Asher Koriat, Joelle Proust,
Nicholas Shea, Valerie Thompson, Maggie Toplak

Chairs:
Balzs Aczl, Olivier Mascaro
Organizers:
Csaba Plh, Nevena Padovan, Dora Kampis
Barnabs Szszi, Aba Szllsi, Bence Plfi

Centre for Advanced Academic Studies (CAAS) Dubrovnik

University of Zagrab

Programme

Thursday, May 21:

17:0019:00 Registration
15:3018:30 Tobii Pro pre-conference workshop
19:00
Welcome Reception

Friday, May 22:

8:009:00
9:0010:00

10:0010:30
10:3011:30
11:3012:30

12:3014:30
14:3015:30
15:3016:00
16:0017:00
17:00

Registration
Invited talk: Valerie Thompson
Intuition, Analytic Thinking, and the Feeling of Rightness: A
Metacognitive Reasoning Theory
Coffee break
Poster session 1
Invited talk: Klaus Fiedler
Metacognitive myopia - a major impediment of rational
behavior
Lunch
Poster session 2
Coffee break
Invited talk: Nicholas Shea
Metacognition of Concepts
Social programme

Saturday, May 23:

9:0010:00

10:0010:30
10:3011:30
11:3012:30
12:3014:30
14:3015:30
15:3016:00
16:0017:00

17:0018:00
18:0018:15
18:1518:30
20:00

Invited talk: Maggie Toplak

The development of rational thinking and decision making
in children and youth
Coffee break
Poster session 3
Paper session (3 presentations)
Lunch
Invited talk: Joelle Proust
Metacognition from a dual-process viewpoint
Coffee break
Invited talk: Asher Koriat
How do we monitor the correctness of our knowledge and
judgments? The bases and accuracy of subjective
confidence
Roundtable
Assembly meeting
Closing words & Farewell reception
Social programme

Friday (May 22), 9:0010:00

Abstracts - Invited speakers

INTUITION, ANALYTIC THINKING, AND THE FEELING OF RIGHTNESS: A

METACOGNITIVE REASONING THEORY

Valerie Thompson,
University of Saskatchewan
Amongst educators and psychologists, questions of metacognition have garnered
much attention, especially as they pertain to the domains of learning and memory. In
this context, it is therefore somewhat surprising that relatively little is known about the
cognitive processes involved in the monitoring and control of reasoning and problemsolving. That is, we know little about how reasoners assess their confidence in their
conclusions, how this confidence influences subsequent behaviour, nor how
reasoning performance is assessed on an ongoing basis. In this talk, I will outline a
theory of metareasoning that draws on insights from the learning and memory
literature, extending them to a variety of common reasoning tasks. Data will be
presented to suggest that monitoring is inferential and based on cues such as the
fluency with which answers are brought to mind. Monitoring is proposed to be a
continuous processes, and the relationship between monitoring and control
processes will be explicated. In addition to corroborating links between the
metatmemory and the nascent field of metareasoning, I will also provide evidence
that study of metareasoning provides novel insights about both metacognition and
reasoning.

Friday (May 22), 11:3012:30

METACOGNITIVE MYOPIA - A MAJOR IMPEDIMENT OF RATIONAL BEHAVIOR

Klaus Fiedler
Ruprecht-Karls-Universitt Heidelberg
What I have come to call meta-cognitive myopia (MM), using a term once
suggested by Robyn Dawes, is the phenomenon that people are pretty accurate in
utilizing even large amounts of stimulus information. However, they are uncritical and
nave regarding the history and validity of the given information samples. This
uncritical reliance on the information given is most conspicuous when the task
context makes it crystal-clear that the stimulus data should not be trusted. I locate
MM within a broader framework of meta-cognition research and illustrate the
phenomenon with examples from various research paradigms. MM offers an
alternative account of many biases in judgment and decision making, which have
been traditionally explained in terms of capacity constraints, limited reasoning ability,
motivational forces, or severely biased environmental input. The explanatory power
of the MM construct, and its theoretical potential to predict new findings, is
demonstrated with reference to five paradigms: inability to discard irrelevant
information; utilization of selectively sampled information; conditional inference
biases; sample-size neglect; and myopia for the impact of aggregation levels. The
final discussion is concerned with the learning origins of MM and the question of why
evolution did not equip homo sapiens with more effective meta-cognitive tools. An
analysis of the costs and benefits will reveal that MM may serve important adaptive
functions, and that eliminating MM may have maladaptive effects. Nevertheless, in
the context of many real decision problems, the costs and irrational consequences of
MM cannot be denied. The final discussion therefore focuses on possible ways to
avoid and alleviate MM and its irrational consequences.

Friday (May 22), 16:0017:00

METACOGNITION OF CONCEPTS

Nicholas Shea
King's College London
Concepts are the constituents of thought and underpin much personal level
reasoning. They also allow us to project properties we have learnt about one object
to new objects. For example, I might interact with something I have classified under
my CAT concept and learn that it purrs when stroked. When subsequently
encountering another object that is classified under CAT I can form the expectation
that it will purr if stroked. Reasoning and projection are two core uses of concepts.
Some concepts are more dependable than others for these purposes. This paper will
suggest that thinkers often make use of a sense of how dependable their concepts
are. Such feelings of dependability are not explicit higher order beliefs about a
concept, but a form of what has been called procedural metacognition (Proust 2013
The Philosophy of Metacognition). Metacognition has been studied in relation to
many cognitive processes, prominently memory and decision making, but it is littlestudied in relation to concepts. This paper makes a prima facie case that there
metacognition of concepts, in the form of a non-conceptual representation or feeling
of dependability that is associated with the use of many concepts. It goes on to
explore some philosophical applications of the idea that there is metacognition of
concepts.

Saturday (May 23), 9:0010:00

THE DEVELOPMENT OF RATIONAL THINKING AND DECISION MAKING IN CHILDREN

AND YOUTH

Maggie Toplak
York University
Many cognitive abilities show a steady increase throughout childhood and
adolescence, but this is not necessarily accompanied by better rational thinking
performance on several heuristics and biases tasks. Previous research has found
that some rational thinking skills show improvement with age, but others do not
(Davidson, 1995; Jacobs & Potenza, 1991; Klaczynski, 2005; Morsanyi & Handley,
2008; Reyna & Farley, 2006; Reyna et al., 2006). Likewise, our research group has
found that some measures of rational thinking performance are associated with
cognitive ability in adults, and some are not (Stanovich & West, 2008). We have used
a taxonomy of rational thinking tasks that predicts which rational thinking tasks will
associate with development and cognitive ability (Stanovich, West, & Toplak, 2011).
Our taxonomy predicts that only tasks necessitating cognitive decoupling and
analytic override will be significantly associated with cognitive ability and
development. We have examined performance on several rational thinking tasks
(probabilistic choice, belief bias syllogisms, resistance to framing, baserate
sensitivity, and otherside thinking) and measures of cognitive ability (intellectual
abilities and executive functions) in a sample of children (N=204) aged 8-14 years of
age (Toplak, West, & Stanovich, 2014). Developmental differences were found on
our rational thinking tasks and associations with cognitive ability were consistent with
these developmental patterns. These findings suggest that rational thinking tasks
which require analytic override show significant correlations with cognitive abilities as
well as developmental trends. This research contributes to clarifying previous
developmental results on heuristics and biases tasks that appeared to suggest
inconsistent developmental trends.

Saturday (May 23), 14:3015:30

THE DEVELOPMENT OF RATIONAL THINKING AND DECISION MAKING IN CHILDREN

AND YOUTH

Jolle Proust
Jean Nicod Institute, CNRS, Paris
There is ample evidence favoring a dual-processing conception of metacognition,
which contrasts Type 1 processes activity-dependent, implicit, fast, automatic, and
inflexible -, with Type 2 processes - activity-independent, explicit, slow, controlled,
and flexible. This contrast, however descriptively adequate, calls for an explanatory
account. The evolutionary pattern of metacognition, the prominent role, in humans, of
epistemic decision at various time scales and resource levels, and the contrast
between associative and inferential cognition point to the engagement of two different
representational systems in human metacognition. While System 1 is based on
relational, subjective, evaluative non-propositional attitudes, System 2 consists in
referential, objective, propositional attitudes, taylored to meet the needs of linguistic
communication. As a consequence of this representational difference, System 1 can
be sensitive to the fluency or effort involved in a cognitive task and predict its likely
correction on this basis. It is not equipped, however, to identify what makes any
specific content true or false. System 2, which can reflectively deal with reasons,
cannot swiftly evaluate or predict correctness of its outputs. To be fully functional,
metacognition must, then, be able to use both systems jointly, in a context-sensitive
way. How is this cooperation at all possible? We will attempt to articulate a tentative
solution to this question, based on a threetiered theory of cognitive action, consonant
with Rolf Reber's recent proposal, in which noetic habits, strategically trained and
monitored, become able to trigger System 1 evaluations.

Saturday (May 23), 16:0017:00

HOW DO WE MONITOR THE CORRECTNESS OF OUR KNOWLEDGE AND

JUDGMENTS? THE BASES AND ACCURACY OF SUBJECTIVE CONFIDENCE
Asher Koriat
University of Haifa
People take the validity of their confidence judgments for granted and use these
judgments to guide their behavior. But how valid are subjective convictions, and what
are the reasons for their validity? A Self-Consistency Model (SCM) will be presented
for the basis of confidence judgments and their accuracy. SCM assumes that the
process underlying subjective confidence has much in common with that underlying
statistical inference about the outside world. Participants behave like intuitive
statisticians who attempt to reach a conclusion about a population on the basis of a
small sample of observations. When asked to decide between two answers, they
retrieve a small sample of clues and considerations from memory, and their
confidence in their decision represents an assessment of reproducibility -- the
likelihood that a new sample of clues will yield the same decision. The results
indicate that metacognitive accuracy is a by-product of cognitive accuracy: The
confidence/accuracy correlation is positive only because this correlation has been
studied under conditions in which people cognitive performance is largely correct,
but is negative when people are largely in error. Results consistent with the model
were obtained across many domains, and the theory has proved yielded novel
predictions for several areas such as the effects of social influence, individual vs.
group decisions, and the wisdom-of-crowd phenomenon.

Saturday (May 23), 11:3012:30

Abstracts Paper session

INDIVIDUAL DIFFERENCES IN STATE-DEPENDENT DECISION MAKING DEPEND ON

COGNITION, AND METACOGNITIVE MONITORING AND CONTROL
Simon A Jackson, University of Sydney, Australia
Sabina Kleitman, University of Sydney, Australia
We carry out or withhold from actions whose utility depends on our judgements being
accurate. Cognitive models suggest that such decisions are carried out when
metacognitive confidence in the presumed state exceeds a decision control criteria.
Little work has been done, however, to determine whether robust individual
differences in decision behaviour can be explained by individual differences in
metacognitive monitoring and control. Undergraduate students (N=364) completed
nine tests assessing fluid and crystallized cognitive abilities, indicated confidence
levels in their answers, and made state-dependent decisions on each answer (e.g.,
submit for marking or not). The nature and quality of individuals' decisions were
described with novel Decision Pattern Analysis metrics: competence, optimality,
recklessness, hesitancy and decisiveness. In line with cognitive theories, the relative
standing of individuals' decision optimality was predicted and manipulated by
changes in the cognitive requirements of the task, while individuals' standings on the
other decision metrics were predicted and manipulated by changes to the nature of
the decision, which influenced individuals' metacognitive variables. Personality
variables made no contribution to the results. The results demonstrate how individual
differences in state-dependent decision making can be explained by differences in
cognition, monitoring and control.
e-mail: simon.jackson@sydney.edu.au

10

Saturday (May 23), 11:3012:30

THE ACCURACY OF METACOGNITIVE JUDGMENTS IN SYLLOGISTIC REASONING

Igor Bajanski, University of Rijeka, Croatia;
Maja Moibob, University of Rijeka, Croatia;
Pavle Valerjev, University of Zadar, Croatia
We carry out or withhold from actions whose utility depends on our judgements being
accurate. Cognitive models suggest that such decisions are carried out when
metacognitive confidence in the presumed state exceeds a decision control criteria.
Little work has been done, however, to determine whether robust individual
differences in decision behaviour can be explained by individual differences in
metacognitive monitoring and control. Undergraduate students (N=364) completed
nine tests assessing fluid and crystallized cognitive abilities, indicated confidence
levels in their answers, and made state-dependent decisions on each answer (e.g.,
submit for marking or not). The nature and quality of individuals' decisions were
described with novel Decision Pattern Analysis metrics: competence, optimality,
recklessness, hesitancy and decisiveness. In line with cognitive theories, the relative
standing of individuals' decision optimality was predicted and manipulated by
changes in the cognitive requirements of the task, while individuals' standings on the
other decision metrics were predicted and manipulated by changes to the nature of
the decision, which influenced individuals' metacognitive variables. Personality
variables made no contribution to the results. The results demonstrate how individual
differences in state-dependent decision making can be explained by differences in
cognition, monitoring and control.
Key words: metacognition, syllogistic reasoning, judgment accuracy, answer
consensuality
Author contact e-mail address: sibajsan@ffri.hr

11

Saturday (May 23), 11:3012:30

DETERMINANTS OF CONFIDENCE IN THE BAT AND BALL PROBLEM

Aba Szollosi, Eotvos Lorand University;
Barnabas Szaszi, Eotvos Lorand University;
Balazs Aczel, Eotvos Lorand University
Decision-confidence differences between conflict and non-conflict versions of several
reasoning tasks gave rise to the claim of an intuitive logical process that proposes
people implicitly detects conflict between the correct and their heuristic answers.
Interestingly, the effect also appeared on the popular Bat and Ball Problem. In two
empirical studies we tested whether implicit conflict detection has any effect on
decision confidence in the Bat and Ball Problem. Verbal measures did not reveal
significant effect of implicit processes between the conflict and non-conflict versions
of the task. A hierarchical regression analysis revealed that the verification of the
operation and the perceived verifiability of the task significantly predicted confidence,
however, the significant predicting power of accuracy raises questions about the
validity of the paradigm.
Key words: metacognition, decision-confidence, CRT, bat and ball problem, conflict
detection
Author contact e-mail address: aba.szollosi@gmail.com

12

Friday (May 22), 10:3011:30

Poster sessions

Poster Session 1.: Friday, May 22, 10:3011:30

Meditative Choice is a Good Choice: Connecting Mindfulness, Decision
Making, and Emotions
Christopher Anderson, Steven Jay Lynn
Better memory improvement after a JOL than after relearning: an effect of
retrieval attempt when monitoring knowledge?
Elisabeth Bacon, Marie Izaute, Elisabeth Bacon
Great apes understand others' beliefs in an interactive task
Frances Buttelmann, Malinda Carpenter, Josep Call, Michael Tomasello
What is the best measure of unpredictability for morphological segmentation?
Daniel Czgel, Csaba Plh
Measuring mindfulness with behavioral assessments
Denke Diana, Monika Albu, Mate Szondy
Interaction between Numbers and Their Quantities for Small Magnitudes
Draen Domijan, Mia eti
The unsmart infant: language comprehension depends on contextual and
situational cues in 9-moth-old infants
Blint Forgcs, Eugenio Parise, Gergely Csibra
The impact of social cues on metacognitive judgments across cultures
Amelie Jacquot, Julie Grzes, Terry Eskenazi, Edith Sales-Vuillemin, Mika Ueno,
Shoji Itakura, Jolle Proust, Laurence Conty
Metacognitive skill-development in Medical
Considering Philosophical Tools of Thinking
Basak Kocaoglu, Cetin Balanuye, Yesim Senol

Education:

Pilot

Study

Tracking social agents mental states modulates action preparedness in adults

gnes Kovcs, Ern Tgls

13

Friday (May 22), 10:3011:30

Self-reported preferred thinking style and cognitive reflection
Eva Ballov Mikukov, Vladimra avojov, Rbert Hank
Congruency sequence effect in the ratio bias paradigm
Bence Plfi, Zoltan Kekecs, Balazs Aczel
Ostensive-Referential Communication Fosters the Interpretation of Pictures of
Objects as Symbols of their own Kind in 9-Month-Old Infants
Eugenio Parise, Gergely Csibra
The effects of script based schemes on retrieval of novel events
Anett Rag, Dvid Szabolcsi, Bence Szekeres-Gal
The effects of valence and images on the source monitoring
Frdrique Robin
How likely is it that Snakes are not Reptiles? A new paradigm approach to
Belief Bias.
Edward Stupple
Enhancing neural excitability of the right vs. left prefrontal cortex differentially
affects consolidation of implicit statistical learning
Csenge Trk, Kata Horvth, Karolina Janacsek, Andrea Antal, Dezso Nemeth
Moral reasoning among adolescents and elderly a group study
Andrea Turner, Judit Pli
In which box does this number go? Breaking down the number categorization
task.
Arnaud Viarouge, Vronique Izard

14

Friday (May 22), 14:3015:30

Poster Session 2.: Friday, May 22, 14:3015:30

An integrative metacognitive model of mindfulness
Mnika Albu, Szondy Mt, Diana Denke
Interference and priming in spontaneous belief computation
Gbor Brdy, Ildik Kirly, gnes Kovcs
Machine Learning for the Estimation of UPDRS score
Krisztina Bza
A Holistic Theory on the Development of Cognitive Biases in Real-Life
Decisions
Marcus Domeier, Pierre Sachse
Psycholinguistic analysis of conjunction in Croatian
Margarita Dragani, Marijan Palmovi
Evidence for spontaneous level-2 perspective taking in adults
Fruzsina Elekes, Mt Varga, Ildik Kirly
Conscious monitoring of emotion perception: evidence from attentional blink
Kamil Fulawka, Remigiusz Szczepanowski
Impact of semantic encoding effects on arithmetic problems in sorting and
solving tasks
Hippolyte Gros, Emmanuel Sander, Jean-Pierre Thibaut
Coordinated Reasoning
Justin Jacot
Is belief re-computation by retrieval available for young children?
Ildik Kirly, Kata Olh, gnes Melinda Kovcs, Gergely Csibra
The domain-generality of working memory: A matter of ability
Kristof Kovacs, Dylan Molenaar, Andrew Conway
"But I was there!" - episodic memory as a mechanism for justification
Johannes Mahr, Reka Peto, Gergely Csibra
Oxytocin modulates social categorization processes
Katalin Olh, Jzsef Topl

15

Friday (May 22), 14:3015:30

Pragmatic aspects of reasoning in real life situations

Marina Oluji, Gordana Hrica, Marijan Palmovi, Marina Oluji, Nevena Padovan
Behavioral task as tool for short-term verbal memory and cognitive
vulnerability to depression assessment in mesial temporal lobe epilepsy
Lidija Preglej, Draen Domijan, Daniel Crnkovi, Ksenija Marinkovi, Hrvoje
Heimovi
Metamemory in educational psychology
Marta Riess
The role of executive functions in patients with aphasia
Izabella Szllsi, gnes Lukcs
A comparison of possession comprehension in children with typical language
development and children with specific language impairments using E-prime
Ivona Radi Tatar, Paulina Mataija, Martina Vukovi Ogrizek, Rafaela Marti, Marina
Paprika
Deductive problem solving and learning logic
Mariusz Urbaski, Katarzyna Paluszkiewicz, Joanna Urbaska
Clinical Gut: The role of inferential direction on
metacognitive confidence and need for more information
Jacinto Sofia, Ferreira Mrio, Braga Joo

16

psychotherapists

Saturday (May 23), 10:3011:30

Poster Session 3.: Saturday, May 23, 10:3011:30

Solvable or Not Solvable? Heuristic Cues that Underlie the Judgment of
Solvability
Rakefet Ackerman, Yael Beller
Assessment of Non-Native Speakers' Metacognitive Abilities and Reasoning
Stem from Rational Thinking to Adopt a New Template of Language
Grammatical Aspects: Evidence from MEG and fMRI
Sadeq Ali Saad Al Yaari
Overcoming intuitive interference in geometry: from brain studies to an
educational intervention
Reuven Babai, Enav Shalev, Ruth Stavy
Contaminated mindware and reflective mind
Vladimira avojov, Rbert Hank, Eva Ballov Mikukov
The impact of social context on children's executive functioning and it's
relationship to creativity
Paula Fischer, Nicolas Chevalier, Mark A. Runco
Representing others' beliefs about object identity in 14 month-old infants
Dora Kampis, gnes Melinda Kovcs
Metacognition and performance rating in spatial manipulation tasks
Roland Kasek, Szabolcs Takcs, Mnika Albu
Negative numbers in elementary school
Orsolya Kiss, Krajcsi Attila
Low cognitive load and reduced arousal impede practice effects on executive
functioning, metacognitive confidence and decision making
Sabina Kleitman, Simon A Jackson, Eugene Aidman
The Source of the Distance Effect in an Artificial Number Sequence
Petia Kojouharova, Attila Krajcsi
The Trusting Twos: The epistemic weight attributed to communication
increases during toddlerhood.
Olivier Mascaro, gnes Melinda Kovcs

17

Saturday (May 23), 10:3011:30

Is writing metarepresentational?
Olivier Morin
Affect and Metacognition: Confidence (in)Sensitivity When Feeling Happy
Yael Sidi, Rakefet Ackerman, Amir Erez
The Roles of Intention and Belief on Mood
Sabina Skubic, Alexander Batthyany
Metacognition at a collective level
Tobias Strzinger
On the cognitive biases as cognitive territories outside positive manifold
continent
Predrag Teovanovi, Lazar Stankov
The development of metacognitive monitoring-control processes in
adolescence
Nike Tsalas, Markus Paulus, Beate Sodian
When problem solving meets metacognition: thinking about writing before
doing it.
Rosa Volpe, Lucile Chanquoy
The Effect of Explicit and Implicit Category Learning on Classification
Accuracy and Confidence
Valnea auhar, Igor Bajanski, Draen Domijan

18

Friday (May 22), 10:3011:30

Abstracts Posters

Session 1
Meditative Choice is a Good Choice: Connecting Mindfulness, Decision
Making, and Emotions
Christopher J. Anderson, Union College, USA
Steven Jay Lynn, Binghamton University, SUNY, USA

Mindfulness has recently shown much promise as an approach to improving physical

and mental health, but little research has examined its role in decision making. We
postulated that dispositional mindfulness and acceptance would be related to both
decisiveness and positive emotions associated with decision making.
Using validated instruments, we tested 68 participants on a battery of relevant
measures: Decisiveness, Regret, the Mindful Attention Awareness Scale, Subjective
Well Being, Honest Responding (BIDR), the Behavioral Symptom Inventory, and an
average Valence of cued autobiographical memories. The data were analyzed by
multiple regression techniques to account for shared variance explained.
Statistically significant connections with strong effect sizes were found in the
predicted domains as well as on some unexpected variables (e.g., people high in
acceptance were very likely to respond honestly, but mindfulness had no relationship
to honesty). The evidence suggests that research into training mindfulness and
acceptance may reveal a effective, strong, and relatively inexpensive route for
improving individual decision making.
Author contact e-mail address: andersc2@union.edu

Better memory improvement after a JOL than after relearning: an effect of

retrieval attempt when monitoring knowledge?
Elin Akdogan1,2; Marie Izaute 3,4; Elisabeth Bacon 1,2,5,6.
1: Unistra (university of Strasbourg, France); 2: INSERM U-1114, Strasbourg; 3: ClermontFerrand Blaise Pascal University, France; 4: CNRS, UMR 6024, LAPSCO, Clermont
Ferrand, France; 5: University Hospital of Strasbourg; 6: Fdration de mdecine
translationnelle, Strasbourg.

The most common learning method focuses on learning repetition, but is not very
effective for long-term retention. More recently, the testing effect shows testing to be

19

Friday (May 22), 10:3011:30

more efficient than re-learning. On the other hand, when making a Judgment Of
learning (JOL), people make a covert attempt to retrieve the answer, which can be
regarded as a self-testing. Studies having explored the potential mnemonic benefit of
repeated JOL have yielded ambiguous results. The aim of this study was to compare
JOL with both testing and re-study conditions in a very plain experimental paradigm.
After a single word-pair encoding phase, participants took part in a single session
devoted to re-study, self-monitoring, or testing. The final cued recall test occurred 48
hours later. The strength of the cue-target association was manipulated. The results
show no differences in performances for the easy word pairs. However, with the
difficult material, the monitoring procedure produced comparable memory
performance to testing procedure, and both yielded better levels of performance than
re-study. Retrieval practice seems to underlie both strategies. JOL could be an
alternative to the testing which could have more chance to be adopted.
Author contact e-mail address: elisabeth.bacon@unistra.fr

Great apes understand others; beliefs in an interactive task

Frances Buttelmann, Central European University, Hungary / University of Erfurt, Germany.
David Buttelmann, University of Erfurt, Germany
Malinda Carpenter, Max Planck Institute for Evolutionary Anthropology, Germany / University
of St Andrews, United Kingdom
Josep Call, Max Planck Institute for Evolutionary Anthropology, Germany / University of St
Andrews, United Kingdom
Michael Tomasello, Max Planck Institute for Evolutionary Anthropology, Germany

Thirty-four great apes theory-of-mind was investigated using an interactive helping

task. An experimenter put an object at location A. An assistant transferred the object
to location B, either in the experimenters absence (false-belief) or presence (truebelief condition). Afterwards the experimenter unsuccessfully tried to open box A.
Subjects were then given the chance to help the experimenter open a box. To help
appropriately, they had to infer the experimenters goal based on his beliefs: In the
false-belief condition, he believed box A to contain his object, thus, he probably
wanted his object. In the true-belief condition, in contrast, the experimenter knew that
box A was empty. Thus, he was less likely to want his object when trying to open the
empty box. Subjects correctly unlocked the box containing the object (box B)
significantly more often in the false-belief than in the true-belief condition. In Study 2,
the same subjects were tested
in a replication of the false-belief and in a new ignorance condition. Again, subjects
unlocked the box containing the object significantly more often in the false-belief than

20

Friday (May 22), 10:3011:30

in the ignorance condition. These studies provide first evidence that great apes
understand others false beliefs, challenging humans uniqueness in this respect.
Author contact e-mail address: visbuttelmann@ceu.hu

What is the best measure of unpredictability for morphological segmentation?

Dniel Czgel, Budapest University of Technology and Economics, Hungary
Csaba Plh, Central European University, Hungary

Morphological segmentation is a crucial part of language processing. It has been

raised intuitively several times that the morpheme boundaries may correspond to a
relative entropy increase in word forms.
A usual approach is to somehow quantify the unpredictability of the next or previous
letter (or phoneme) in the text. An important point here is how to choose this
unpredictability measure. In the literature, three main methods have been proposed:
letter successor/predecessor variety (LSV/LPV), Shannon entropy (LSE/LPE), and
max-drop (LSM/LPM).
In our work, all these three measures are generalized by using a one-parameter
entropy measure called Rnyi-entropy S(q). Based on S(q), morpheme boundaries
are inferred in an unsupervised manner from the text. In order to investigate what
unpredictibility measure, parametrized by q, is the most applicable for this task, we
evaluate our results by comparing it with grammatically segmented data from the
Hungarian National Corpus containing about 200 million words.
Author contact e-mail address: czegel_d@yahoo.com

Measuring mindfulness with behavioral assessments

Diana Denke, KRE - Psychological Institue, Budapest, Hungary
Monika Albu, KRE - Psychological Institue, Budapest, Hungary
Mate Szondy KRE - Psychological Institue, Budapest, Hungary

Mindfulness techniques and related research have become more and more
widespread in recent years. Most studies use self-report questionnaires to assess
dispositional capacities of mindfulness and examine the efficiency of mindfulnessbased trainings in the light of correlating psychological factors (mental health,
emotional difficulties, and stress reactions). However validated behavioral
assessments that would objectively measure mindfulness capabilities are rare to
found. In our study two behavioral tests were developed for the above purpose.

21

Friday (May 22), 10:3011:30

First, results from a breath counting exercise have been found correlating with
dispositional mindfulness (FFMQ, Baer et al. 2006) and separate from various
indicators of working memory. Second, an exercise based on the Necker cube
illusion was used as a behavioral assessment to measures cognitive flexibility and
conscious presence. Various conditions of the Necker cube exercise have been
found correlating with FFMQ variables and partly with indicators of working memory
and executive functions.
Results of the study indicate that breath counting and Necker cube exercises could
serve as behavioral tests to objectively measure mindfulness capacities.
Key words: Mindfulness, behavioral assessments, FFMQ
Author contact e-mail address: diana.denke@gmail.com

Interaction between Numbers and Their Quantities for Small Magnitudes

Draen Domijan, University of Rijeka, Croatia
Mia eti, Catholic University of Croatia, Croatia

In two experiments, we examined whether irrelevant numerical information can

influence the speed of sentence-picture verification. Participants were asked to verify
whether the concept mentioned in a sentence matched the shape of the object
presented in a subsequent picture (concept-object match). Concurrently, the number
attached to the concept in the sentence and the quantity of objects presented in the
picture was manipulated (number-quantity match). The number of objects varied from
one to four. In experiment 1, sixty-four statements were created: 32 statements which
matched in shape and 32 did not match in shape with the subsequent picture
presentation. Within each set, half of the statements also matched number with
quantity and the other half did not, creating a 2 x 2 factorial design with conceptobject match (yes vs. no) and number-quantity match (yes vs. no) as repeated
measures factors. In experiment 2, the stimuli were full sentences. In both
experiments, the verification times for concept-object match were faster when there
was also a number-quantity match compared to mismatch. On the other hand, there
was no difference between number-quantity match and mismatch when concept and
object mismatched.
Key words: grounded cognition, numerical cognition, sentence-picture verification,
Stroop interference
Author contact e-mail address: ddomijan@ffri.hr

22

Friday (May 22), 10:3011:30

The unsmart infant: language comprehension depends on contextual and

situational cues in 9-moth -old infants
Blint Forgcs, Universit Paris Descartes
Eugenio Parise, Lancaster University
Gergely Csibra, Central European University

A number of electrophysiological studies have reported that infants as young as 9

months of age exhibit the neural correlate of semantic comprehension, the N400
ERP component. These results seem to confirm the long-standing assumption that
language comprehension is in place prior to language production. However, infants in
this age seem to detect semantic incongruity only under specific conditions, for
example, if words are produced by their mother (and not by an experimenter), or if
they are engaged in an ostensive-referential live interaction. Such results suggest
that 9-month-olds might understand words only if they hear them from a familiar
voice, or if they are highly engaged. In an EEG experiment using a puppet theater
presentation we aimed to explore whether infants understand words if they do not
hear them from their mother, but merely if they are highly engaged. Our results
brought no N400 effects, which indicates that 9-month-old infants might not treat
language independent of situational and contextual aspects, and the identity of the
speaker, or that phonological facilitation (the mother's voice) might be decisive in
their comprehension capacities.
Key words: word acquisiton, N400, infant development, language comprehension
Author contact e-mail address: balint.forgacs@parisdescartes.fr

The impact of social cues on metacognitive judgments across cultures

Amelie Jacquot, LPN, University of Paris 8, France
Julie Grzes, LNC, ENS Paris, France
Terry Eskenazi, LNC, ENS Paris, France
Edith Sales-Vuillemin, SPMS, University of Dijon, France
Mika Ueno, University of Kyoto, Japan
Shoji Itakura, University of Kyoto, Japan
Jolle Proust, IJN, ENS Paris, France
Laurence Conty, LPN, University of Paris 8, France

Metacognitive judgments refer to processes by which people assess their own

cognitive operations. We showed that these judgments are susceptible to unreliable
non-verbal social signals. Here, we tested whether such susceptibility varies as a
function of individual's culture. Indeed, eastern cultures are usually considered as
more collectivist than western cultures. Moreover, it has been robustly reported that
people are more sensitive to information provided by ingroup members. Thus, we
addressed whether people are more sensitive to information provided by their own

23

Friday (May 22), 10:3011:30

ethnic group and whether eastern participants are more sensitive to information
shared by several sources. We realized a cross-cultural study (France/Japan) where
participants were confronted with social information after having performed a first
order task. Social information consisted in faces that turned their attention toward the
participant's response (congruent trials) or toward the opposite re
sponse (incongruent trials). At each trial, the participants were asked to rate their
confidence in their response. We manipulate the number of social sources and the
group membership. Preliminary results indicated that the participants' confidence
raised for congruent as compared to incongruent trials. Moreover, this effect
increased with the number of social sources. Ongoing analyses will reveal whether
culture and group membership impact these effects.
Key words: metacognition, social influence, cross-cultural
Author contact e-mail address: amelijacquot@gmail.com

Metacognitive skill-development in Medical Education: A Pilot Study

Considering Philosophical Tools of Thinking
Basak Kocaoglu, Akdeniz University, Turkey
Cetin Balanuye, Akdeniz University, Turkey
Yesim Senol, Akdeniz University, Turkey

Philosophical reasoning is a slightly distinct way of thinking. It has specific features

such as examination of thought systems by thought itself, and manipulation of
thought processes. However, metacognition is a term that refers to awareness of
ones thought processes and ability to control them. We hypothesized that the nexus
between philosophical strategies of thinking (PST) and metacognitive skills (MS) is
essential for decision making in critical situations, especially in medicine. In fact, both
studying and practicing medicine have many difficulties such as retrieving of
particular data from medical data stack in a particular case and inferring the
adequate consequences.
The aim of our research is, basically, considering these strategies, whether
improvement is possible or possible to what extent in MS. We designed pilotcurriculum to a group of second-term medical students (n=10). This scheme is
composed with basic PST, for instance, hypothetico-deductive method, exposing
causality with necessary and sufficient reasons. These strategies which interwoven
into second-term medical education, taught in a selective course (2 hours per week).
This study is an ongoing project that due date is May 7th, 2015. Students hitherto
have been assigned two tasks. No significant relationship was found between gender
and MAI scores (p>0.05).

24

Friday (May 22), 10:3011:30

Tracking social agents mental states modulates action preparedness in adults

gnes Melinda Kovcs, Central European University, Hungary
Ern Tgls, Central European University, Hungary

Successful social interactions require understanding that representations about the

world reflect a subjective point of view. However, the knowledge a person brings into
the interaction may be inaccurate and sometimes even incorrect. Consequently,
these expectations are sometimes driven by false beliefs. Recent studies suggest
that 18-month-olds not only track a protagonists belief, but show active helping
(Buttelman et al., 2009) or attempts to correct the possibly erroneous actions
(Knudsen & Liszkowski, 2012) if they expect that others' behavior will be driven by
false beliefs. In the present study we investigated whether processing others mental
states will elicit spontaneous preparedness for action also in adults. Participants
(N=25) watched movies involving a central character that placed a ball in a pipesystem with opaque segments. In the false belief condition the ball moves to a new
location while the character is turned away. In the true belief condition the character
sees this transfer. Participants were asked to continuously press a button till a
specific cue was presented, and then point to the actual location of the ball on a
touchscreen. Our data show that participants released the button faster in the false
than in the true belief condition, suggesting that tracking other agents false beliefs
results in a greater preparedness for action in adults.
Author contact e-mail address: teglase@ceu.edu

Self-reported preferred thinking style and cognitive reflection

Eva Ballov Mikukov, Vladimra avojov, Rbert Hank,
Institute of Experimental Psychology, Slovak Academy of Sciences, Slovakia

The aim of the study was examination of relationship between self-reported preferred
thinking style and cognitive reflection.
Total sample of 275 participants (225 women) completed the Cognitive reflection test
(CRT, Frederick, 2005) - 7 items version measuring ability to override intuitive
response and find a correct answer and the Type of intuition scale (TIntS, Pretz et
al., 2014) measuring self-reported preference for four intuitive thinking style: holisticabstract intuition, holistic-big picture intuition, inferential intuition and affective
intuition. 98 of participants (78 women) completed also the Rational-experiential
inventory (REI, Pacini & Epstein, 1999) measuring self-reported preference for
rational and experiential information processing.

25

Friday (May 22), 10:3011:30

Correct answers in CRT correlated positively, but weak, with rational thinking style
(r=.301,p=.003) and inferential intuition (r=.132, p=.029); and intuitive answers in
CRT positively correlated with experiential thinking style (r=.262, p=.009) and
affective intuition (r=.151,p=.012). On the other hand, positive correlations were
found between self-reported scales REI and TIntS (for REI-R TIntS-I
r=.264,p=.009; REI-E TIntS-A r=.602,p<.000; REI-E TIntS-HA r=.397,p<.000,
REI-E TIntS-I r=.288,p=.004). Weak and moderate correlations between selfreported preference for intuitive and rational thinking style, but only weak
relationships of self-reported preferences with performance in CRT point out that
people were consistent in self-assessment, but the performance did not meet selfimage.
This work is part of research project Centre of Excellence SAS 'Centre for Strategic
Analyses "(CESTA) III / 2 /, 2011.
Key words: inuition, cognitive reflecton, thinking styles, rationality
Author contact e-mail address: expsebal@savba.sk

Congruency sequence effect in the ratio bias paradigm

Bence Palfi, Eotvos Lorand University, Budapest, Hungary
Zoltan Kekecs, Baylor University, Department of Psychology and Neuroscience, USA
Balazs Aczel, Eotvos Lorand University, Budapest, Hungary

The ratio bias is a well-known task in the reasoning paradigm where people tend to
experience conflict between the heuristic and analytic answers (Denes-Raj &
Epstein, 1994) or between two possible choices (Bonner & Newell, 2010). The
intervention of top-down processes are crucial in resolving this conflict. However, the
recruitment the cognitive control is still an open question. The conflict monitoring
hypothesis (Botvinick, Braver, Barch, Carter & Cohen, 2001) suggests that the
experience of conflict plays a crucial role in the activation of the top-down processes.
In this research, we tested the conflict monitoring hypothesis in the ratio bias
paradigm. After controlling for possible feature repetitions effects, our analysis
revealed that the experience of the conflict results in a better subsequent conflict
resolution. Limitations and future directions are discussed.
Key words: Reasoning, ratio bias, congruency sequence effect, conflict adaptation,
cognitive control
Author contact e-mail address: palfibence@gmail.com

26

Friday (May 22), 10:3011:30

Ostensive-Referential Communication Fosters the Interpretation of Pictures of

Objects as Symbols of their own Kind in 9-Month-Old Infants
Eugenio Parise, Lancaster University, UK
Gergely Csibra, Central European University, Hungary

We addressed the question whether ostensive-referential communication could make

infants take familiar objects as exemplars of their own kind. We measured the P300
ERP component and desynchronisation of alpha-band EEG oscillations in a category
oddball paradigm.
Experiment 1: 9-month-old infants were presented with exemplars of mugs and
spoons in an oddball fashion. Participants responded to the infrequent stimuli with
larger P300 component over posterior region, but no attenuation of alpha oscillation
was found. This suggests the infants perceived the oddball objects, but did not
identify them as belonging to a different kind.
Experiment 2: 9-month-old infants were exposed to the same events as in
Experiment 1, but each trial was preceded by an ostensive utterance and a pointing
hand. A larger P300, as well as stronger alpha attenuation were found in response to
the oddball category suggesting that ostensive-referential communication fosters the
interpretation of objects as exemplars of their own kind.
Infant-directed ostensive communication could designate an object as a symbol for
its own kind, which may support the learning of generic object properties.
Author contact e-mail address: eugenioparise@tiscali.it

The effects of script based schemes on retrieval of novel events

Anett Rag , University of Etvs Lorand, Budapest
Dvid Szabolcsi, University of Etvs Lorand, Budapest
Bence Szekeres-Gal , University of Etvs Lorand, Budapest

Our study of semantic memory organization examined schemes' effect to retrieval.

Based on Schank's dynamic memory theory we tried to activate participants' TOPs,
and therefore influence the retrieval of their event memories.
By reediting different movie trailers we created novel events that were incoherent
(with incomprehensible storyline). We also manipulated the mode of schema
activation by either naming the trailer's genre (labeling) or presenting priming
stimulus beforehand. In the control condition no schema was activated.
We hypothesized that viewers will selectively retrieve some part of the trailer
according to their own expectations when they meet an incomprehensible storyline.
However, if we actively manipulate scheme activation either with explicitly labeling in

27

Friday (May 22), 10:3011:30

advance, or implicitly (priming) can influence retrieval. The more coherent the
narrative, the more accurate is remembering.
We measured retrieval by rating the coherence of narratives given by the
praticipants. Results show the priming group's stories were significantly less coherent
than those of both the label and the control groups. Also the label group performed
significantly better than the other two groups in recollection of the presented stimuli.
We assume that retrieval of novel events is fairly accurate by itself but can be spoiled
by using an inadequate scheme.
Key words: semantic memory, script theory, coherence, retrieval
Author contact e-mail address: carterroulier@gmail.com

The effects of valence and images on the source monitoring

Frdrique Robin, University of Nantes, France.

The pictures recognition depends on the recovery of perceptual and semantic clues
via a recollection processes. The words recognition will instead be based on a
familiarity process, the phonemic and the lexical clues. However, with an imaging
instruction, the representation of the word then contains a quasi sensory-perceptual
attribute which increases likelihood that recognition will be based on the recollection.
But, in this case, it should be more difficult to decide whether the studied item was a
word or a picture. Nevertheless, the emotional valence of the items could reinforce
the distinction between words and pictures. A corpus of words and photos (positive
valence, negative and neutral) was established for the study. One week after seeing
photos with words or words alone, participants have to clarify whether the word has
been seen with a photo or has been imagined. Results show that words are better
recognized when they are presented with photos. In contrast, the fal
se recognitions rates are higher when words have been imagined. Lastly, the
recollection based on details is more frequent when items are positive and when they
have been perceived. Take into account the emotional valence of the information and
its format are important factors which allow us to move forward on processes
involved in false memories creation, for example in the eyewitness testimony.
Key words: emotion, imagination, false memory, source monitoring
Author contact e-mail address: frederique.robin@univ-nantes.fr

28

Friday (May 22), 10:3011:30

How likely is it that Snakes are not Reptiles? A new paradigm approach to
Belief Bias.
Edward J. N. Stupple, Amy Benbow-Hebbert
University of Derby
The new paradigm in reasoning research emphasises a shift away from dichotomous
true/false statements and instead embraces the probabilistic nature of the real world.
In belief-bias studies, belief and logic are manipulated to induce heuristic-analytic
biases, and participants are asked if conclusions necessarily follow from premises.
The present study replicated previous belief bias experiments but instead asked how
likely conclusions were to follow, hypothesising that response patterns and response
times would replicate the dichotomous response studies. In an online experiment 44
participants (13 male participants and 31 female) responded on a sliding scale
whereby 10 = necessarily true, 5 = possibly true and 0 = impossible. Standard effects
of belief-bias were replicated, with main effects of Validity (p<.001) and Belief
(p<.001) and an interaction between the two (p=.036). Moreover, response time
effects replicated previously demonstrated patterns, with main effects of validity
(p<.001) demonstrating increased response times for invalid conclusions and with
significant interaction (p<.001) whereby invalid-believable problems exhibited the
longest response times. These data indicate that current theories of belief bias are
robust to a shift toward a probabilistic variation of the belief bias paradigm and may
be adapted to incorporate probabilistic variations of the task.
Key words: Belief-bias, Dual Process theory, Reasoning
Author contact e-mail address: e.j.n.stupple@derby.ac.uk

Enhancing neural excitability of the right vs. left prefrontal cortex differentially
affects consolidation of implicit statistical learning
Csenge Trk, Institute of Psychology, Etvs Lornd University, Budapest, Hungary
Kata Horvth, Institute of Psychology, Etvs Lornd University, Budapest, Hungary
Karolina Janacsek, Institute of Psychology, Etvs Lornd University, Budapest, Hungary
Andrea Antal, Institute of Medical Psychology and Medical Sociology, University Medical
Center, Georg-August University, Gttingen, Germany
Dezso Nemeth, Institute of Psychology, Etvs Lornd University, Budapest, Hungary

Picking up statistical regularities of patterns from the environment is essential for

predictive and adaptive behavior, decision making and reasoning. One of the most
important challenges is to understand how statistical learning occurs and how the
acquired information consolidates and stabilizes in the brain. Evidence suggests that
the prefrontal cortex (PFC) has a critical role in these processes; the division of labor
between hemispheres, however, is less characterized. The aim of the present study

29

Friday (May 22), 10:3011:30

was to directly investigate the causal role of the right and left PFC in
implicit/incidental statistical learning and its consolidation. Transcranial direct current
stimulation (tDCS) over the right or left dorsolateral PFC (DLPFC) was applied during
implicit statistical learning in order to modify learning-related cortical plasticity in the
targeted brain regions by increasing neural excitability. Performance was tested
during stimulation and 12-hour later. In order to investigate how brain stimulation
during learning affects sleep-dependent memory consolidation, we compared
experimental groups who did or did not have sleep in the 12-hour consolidation
period (AM-PM vs. PM-AM design). Here we show no sleep effect on consolidation of
implicit sequence memories in sham condition, however, the interaction between
sleep/no sleep condition and tDCS on right vs. left DLPFC was significant. Our
results can lead to a deeper understanding of the relationship between sleep and
memory consolidation.
Key words: implicit learning, transcranial direct current stimulation (tDCS), sleep,
statistical learning and consolidation, prefrontal cortex
Author contact e-mail address: csenge.torok@gmail.com

Moral reasoning among adolescents and elderly- a group study

Andrea TURNER - Kroli Gspr University of the Reformed Church, Budapest, Hungary
Judit Pli PhD - Kroli Gspr University of the Reformed Church, Budapest, Hungary

OBJECTIVES: Kohlberg defined moral reasoning as a thinking process with the

objective of determining whether an idea is right or wrong. His theory specifies six
stages of moral development, arranged in three levels where individuals can only
progress through stages one stage at a time and can neither skip stages nor return to
any previous stage. The objective of this study was to examine the development of
moral reasoning in two age groups.
METHOD: A group study was designed by interviewing subjects about particular
hypothetical dilemmas asking for solutions and a justification for that solution in two
age groups of similar socio-economical status: adolescents (15 girls, 14 boys, mean
age: 17,45 yrs, SD: 0,68) and elderly (15 men, 17 women, mean age: 72 yrs, SD:
4,83).
RESULTS: Significant difference was found between groups with respect to the
principled morality index and also the co-existence of more than two moral stages
could be observed within each group.
CONCLUSIONS: Our findings support the theory of the Neo-kohlbergian complex
model of moral development where a subject is not assigned one for one to a stage
over development but stages overlap and mix in a complicated way.
Key words: Kohlberg, moral reasoning, Neo-kohlbergian model, principled morality
Author contact e-mail address: aturner@t-online.hu

30

Friday (May 22), 10:3011:30

In which box does this number go? Breaking down the number categorization
task.
Arnaud Viarouge, CNRS / Universit Paris Descartes, Paris, France
Vronique Izard, CNRS / Universit Paris Descartes, Paris, France

During early school years, children develop an ability to represent the constant
distance separating consecutive integers. To assess this, the most commonly used
task consists in asking children to indicate the location of a number on a physical line
representing a given numerical interval. An alternative, more intuitive task consists in
asking the participants to categorize numbers according to their size. Both paradigms
have shown a shift in childrens responding with age, going from a compressed
(logarithmic-like) pattern, where more numbers are placed on the side of large
numbers, to a linear pattern, where numbers are evenly placed on the line or across
the categories. We tested a group of 362 participants between 1st and 3rd Grade on
a number categorization task using three categories (small, medium, large) to
divide the 0-100 interval. While we replicate the developmental trajectory described
above, detailed regression analyses showed ambiguous response patterns in many
participants, some of them even producing apparent inverse-log scales. Our results
question the validity of the numerical categorization task as a tool to assess
childrens scale of representation of numbers, as the specific strategies used by
children in this paradigm may preclude access to their mental scale for number.
Key words: numerical cognition, categorization, development, number representation
Author contact e-mail address: arnaud.viarouge@parisdescartes.fr

31

Friday (May 22), 14:3015:30

Session 2
An integrative metacognitive model of mindfulness
Mnika Albu, KRE - Psychological Institute, Budapest
Szondy Mt, KRE - Psychological Institute, Budapest
Diana Denke, KRE - Psychological Institute, Budapest

Mindfulness is often defined as a mental state achieved by focusing ones awareness

on the present moment and acknowledging ones feelings, thoughts and bodily
sensations. Kabat-Zinn (1990) describes mindfulness as bringing attention to
moment-to-moment experience.
Here we provide a theoretical framework and systems-based multi-level
metacognitive model of mindfulness that focuses on integration of top-down and
bottom-up forms of self-regulation. We postulate and discuss following hypothesis:
(1) mindfulness is related to the highest level of metacognition; (2) mindfulness
depends on dynamic cooperation of three main components of the metacognition
(metacognitive knowledge, metacognitive experiences and metacognitive skills); (4)
intentionally practiced mindfulness decreases dissociations between awareness and
meta-awareness; (5) components of mindful meta-level develop and change during
continuous practice. The model is discussed in the light of other theoretical
approaches to mindfulness concept and empirical data provided by studies
examining the relations between mindfulness, hot and cold executive functions and
metacognitive abilities.
We believe that presented model provides some helpful guidelines for future basic
and clinical research, specifically targeting areas of development in the treatment of
stress-mediated psychological disorders.
Key words: metacognition, mindfulness, meta-awareness, compassion
Author contact e-mail address: albumoni@gmail.com

Interference and priming in spontaneous belief computation

Gbor Brdy, Central European University, Hungary
Ildik Kirly, Etvs Lornd University, Hungary
gnes Kovcs, Central European University, Hungary

Humans encode other agents mental states and these attributed mental contents
affect participants behavior. Kovcs et al. (2010) found that the attributed belief
about an object being behind the occluder primed the participants response in a

32

Friday (May 22), 14:3015:30

visual detection paradigm resulting in faster detection of unexpected objects. Will the
avatars belief interfere with the subjects resulting in slower object detection if there
are different beliefs about what object is behind the occluder? In the current study,
with the avatar being present one of the objects moved behind the occluder. Then
this object is switched for the other; this switch is either witnessed by the avatar or
not, depending on the condition (True/False belief). Then the occluder is raised, and
one of three outcomes are presented for the participant: one of the two or both
objects; participants reaction time being measured in detecting the outcome. For the
one object outcomes we found that the participants (N=22) had faster RT for objects
they expected (effect of own expectation (p=.02)). Crucially we also found that they
were slower when the avatar had a different belief (effect of belief condition(p=.03)).
This result shows that attributed beliefs can interfere with participants own beliefs.
Author contact e-mail address: gaborbrody@gmail.com

Machine Learning for the Estimation of UPDRS score

Krisztian Buza, BioIntelligence Lab, Institute of Genomic Medicine and Rare Disorders,
Semmelweis University, Hungary

Parkinson's disease is one of the most remarkable neurodegenerative disorders

affecting motor and cognitive abilities of elder people. One of the key components of
successful treatment is regular monitoring of the patient's status. Usually, UPDRS
score is used for regular monitoring which is often time consuming and it is highly
affected by examinator subjectivity. In order to continuously measure the patient's
UPDRS score, telemonitoring was proposed. The basic idea is to estimate UPDRS
score based on biomedical voice recordings that can be captured while the patient
makes telephone calls or Skype calls using his/her smartphone or tablet.
In our work, we aim to estimate UPDRS score based on biomedical voice recordings
using regression methods. We compare various regression techniques including
linear regression, regression trees, neural networks and nearest neighbor regression.
Our results show that the best regression technique is able to achieve mean absolute
error of 3.08 on predicting the motor UPDRS scores in the publicly avilable
Parkinsons Telemonitoring dataset from the UCI repository if the initial UPDRS score
of the patient is taken into account. This evaluation protocol simulates the scenario
when the UPDRS score is initially measured by a clinician and subsequently the
UPDRS score is estimated in a telemonitoring setting.
Author contact e-mail address: buza@biointelligence.hu

33

Friday (May 22), 14:3015:30

A Holistic Theory on the Development of Cognitive Biases in Real-Life

Decisions
Markus Domeier, University of Innsbruck, Austria
Pierre Sachse, University of Innsbruck, Austria

Real-life decisions, contrary to decisions in the laboratory, mostly take place in

complex situations. Thus, it&#8217;s hard to find the exact reasons, which lead to
sub-optimal decisions. Therefore, this paper analyses the interaction between the
characteristics of complex situations (Schroda, 2000), the computational architecture
of psychological processes (PSI theory, Drner, 2013) and the occurrence of
cognitive biases (Carter, Kaufmann, Michel, 2007).
The developed model delivers a) an approach to identify the situation's
characteristics (complexity, dynamics, time pressure, uncertainty, etc.) which trigger
motivational, emotional and cognitive influences on the decision-making process and
b) an explanation which combination of these influences lead to which kind of
cognitive bias.
In a first step, decision experts rated the influence between the components of the
PSI theory and the different cognitive biases with a cross-impact-matrix (Vester,
1994). The second step aims at verifying the findings in the field by analyzing
erroneous decisions of entrepreneurs. According to Shepherd, Williams and Patzelt
(2015), they are more biased in decision-making than non-entrepreneurs.
The holistic approach provides a deeper insight into the network of cause and effect
of the development of cognitive biases. It therefore serves as a good starting point for
a focused debiasing intervention.
Key words: Cognitive Bias, Decision-Making, Complexity, PSI theory, Cognitive
Architecture
Author contact e-mail address: markus.domeier@gmail.com

Psycholinguistic analysis of conjunction in Croatian

Margarita Dragani, University of Zagreb, Croatia
Marijan Palmovi, University of Zagreb, Croatia

Agreement patterns vary significantly across languages and constitute a challenge

for linguistic theories. The obligatory nature of agreement in Slavic languages makes
them felicitous for studying the factors that contribute to the computation of
agreement patterns. In this study the preferences for agreement are studied with
eye-tracking (regressions and fixations within AOI) in sentences in which the Subject
NP is a conjunction of words of different gender and the Predicate is a Participle.

34

Friday (May 22), 14:3015:30

Croatian speakers prefer default to linear agreement, i.e. in sentences in which the
Subject is a conjunction of feminine and neuter nouns the preference for the
Predicate will be the Participle in masculine. Theoretical accounts of this findings will
be given in terms of Boolean IP constituents.
Key words: agreement, conjunction, gender, number, Croatian
Author contact e-mail address: mddraganic@gmail.com

Evidence for spontaneous level-2 perspective taking in adults

Fruzsina Elekes, Etvs Lornd University, Budapest, Hungary; Central European
University, Budapest, Hungary
Mt Varga, Budapest University of Technology and Economics, Budapest, Hungary
Ildik Kirly, Etvs Lornd University, Budapest, Hungary; Central European University,
Budapest, Hungary

Spontaneous, online perspective taking seems to be limited to tracking which objects

others have visual access to (level-1). Representing how those objects appear to
them (level-2) has not been found in indirect tests previously. We hypothesized that
spontaneous level-2 perspective taking would occur given that the two participants
had the same task on jointly attended stimuli. This creates a mutual "problem field",
ensuring that participants had the same object feature in their focus of attention.
Pairs of participants sat opposite each other and saw symmetric (0,8) and
asymmetric (6,9) stimuli on a screen, laid between them. Participants either had the
same task (number verification) or different tasks (number verification and n-back
based on the color of the characters). They had both individual and joint trials. In the
same task group, jointness worsened RT-s for the asymmetric numbers more than it
affected RT-s for symmetric numbers. In the different tasks
group no such interaction was found. Our findings reveal that the phenomenon of
on-line level-2 perspective taking exists. Results indicate that sharing a problem field
triggers involuntary perspective taking, but being simultaneously engaged with the
stimuli by performing different tasks is not sufficient to do so.
Author contact e-mail address: elekes.fr@gmail.com

35

Friday (May 22), 14:3015:30

Conscious monitoring of emotion perception: evidence from attentional blink

Kamil Fulawka, University of Social Sciences and Humanities Faculty in Wroclaw, Poland
Zhao Fan, Central China Normal University, China
Remigiusz Szczepanowski, University of Social Sciences and Humanities Faculty in
Wroclaw, Poland

According to a higher-order approach, metacognition is the underlying mechanism of

conscious awareness. One of the important functions of metacognition is to monitor
the state of first-order representations via a bottom-up fashion. In the present study,
we investigated bottom-up monitoring of emotional responses by examining a
hypothesis of the relationship between emotion discrimination and metacognitive
content regarding the correctness of decisions under the attentional blink paradigm.
Participants were presented with happy or fearful expressions within the stream of
rapidly changing distractors (RSVP), and rated their metacognitive contents with
confidence ratings regarding accuracy of perceptual discrimination. Similarly to other
attentional blink studies on emotion perception our results showed above-chance
perceptual discriminations for both emotional expressions, although stronger
responses were shown for fearful items. We also observed that participants were
aware of the correctness of first-order decisions, however metacognitive contents in
higher extent reflected perceiving fearful targets. Our results have implications for
higher-order theories of consciousness suggesting that information of greater
biological relevance generate more vivid first-order representations resulting in
stronger higher-order representations through bottom-up conscious monitoring.
ACKNOWLEDGMENTS. This research has been supported by the National Science
Center (Poland), and funded under the grants decision DEC-2011/03/B/HS6/01799
to R.S.
Key words: Metacognition, Attentional Blink, Emotions, Facial Expressions
Author contact e-mail address: kamilfulawka@gmail.com

Impact of semantic encoding effects on arithmetic problems in sorting and

solving tasks.
Hippolyte Gros, University Paris 8, France
Emmanuel Sander, University Paris 8, France
Jean-Pierre Thibaut, University of Burgundy, France

In the process of understanding a mathematical problem, representations are

induced by the solver at an intermediate level of abstraction that neither coincides
with a situation model nor with the abstract mathematical structure. In two

36

Friday (May 22), 14:3015:30

experiments, we manipulated the nature of the quantities (e.g., time units, distance
units, monetary units, etc.) that could mediate these representations.
We created simple arithmetic problems that could all be solved by two strategies
(either a one-step strategy or a three-step strategy). We hypothesized that only one
strategy would be available at a time, depending on the semantic representation
induced.
In the first experiment, we asked 120 adults to sort problems depending on the
strategies they would use to solve them. A multidimensional scaling analysis
performed on the resulting categories showed that the main factor accounting for the
spontaneous classifications was consistent with the variations introduced between
the quantities.
In the second experiment, 60 participants were instructed to solve similar problems
using a minimal number of steps. The results showed that the choice of one of the
two possible solving strategies depended on the representations induced by the
problems, even when the subjects were instructed to use the shortest strategy they
could think of.
Key words: reasoning, problem solving, semantic structure, strategy choice
Author contact e-mail address: hippolyte.gros@gmail.com

Coordinated Reasoning
Justine Jacot, University of Lund, Sweden

When trying to explain how humans reason, it is common to look for applications of
inference patterns, such as deductive laws of logic. Deductive reasoning tasks focus
on testing the way lay people use some logical skills, taking for granted that the 'logic'
used is classical logic. However, if one agrees that one must reason 'to' an
interpretation before reasoning 'from' an interpretation, coordination on the meaning
of instructions between experimenters and subjects is a precondition to drawing
conclusions about reasoning. Moreover, the linguistic form of the instructions yield
certain pragmatic constraints on the semantic content of the instructions, so that the
relevant information is not automatically identified by subjects. Through the example
of the Double Disjunction Task used in the Theory of Mental Models, I propose a new
model for reasoning tasks, based on coordination between experimenters and
subjects, not only on the syntactic form of the reasoning, or the semantic content of
the instructions, but also on the pragmatic influences that constrain those
instructions. This model can be represented as a double signaling game between the
subject and the experimenter, where a solution exists when coordination is reached
at several levels in each game and across games.
Key words: Psychology of reasoning, logical inference, pragmatics
Author contact e-mail address: justine.jacot@fil.lu.se

37

Friday (May 22), 14:3015:30

Is belief re-computation by retrieval available for young children?

Ildik Kirly, Etvs Lornd University, Hungary
Kata Olh, Etvs Lornd University, Hungary
gnes Melinda Kovcs, Central European University, Hungary
Gergely Csibra, Central European University, Hungary

A newly emerging question in Theory of Mind (TOM) research is when exactly the
attributed belief is computed. Taking as example the standard location change task
(SLCT), one can assume that belief is computed when (i) Sally puts her chocolate
into box A, or (ii) when Sally comes back and is going to choose from the containers
in order to find her chocolate. The second possibility (ii) also includes explicit belief
updating situations that could occur when the task itself or other situational
requirements induce a memory search process to find information related to the
belief (as a potential cause to predict behavior) of a social partner. We tested 3-yearold children with a modified SLCT in which children were presented a true belief
situation, in which the model was present during the location change and was
wearing sunglasses. However, then children were allowed to explore her sunglasses,
and it turned out that the sunglasses were opaque. Thus based on their novel
knowledge on the situation children could infer that the model could not follow the
situation despite her presence. If children can re-compute the belief content of the
model, they should act as in a false belief situation. Preliminary results suggest that
3-year-olds can re-calculate the attributed belief based on their memories of earlier
events.
Author contact e-mail address: kiraly.keszei.ildiko@gmail.com

The domain-generality of working memory: A matter of ability

Kristof Kovacs, Eszterhazy Karoly College, Hungary
Dylan Molenaar, University of Amsterdam, The Netherlands
Andrew Conway, Princeton University, USA

Working memory refers to the set of processes that enable one to hold goal-relevant
information in mind in the face of concurrent processing and/or distraction.
Individual differences in working memory are studied with complex span tasks, which
belong to different domains. Latent variable analysis reveals a general factor of
working memory, which is generally referred to as working memory capacity or
WMC. Thus, whereas working memory as an intra-individual contruct is domainspecific, WMC appears to be largely domain-general. Yet this conclusion is
controversial: there are latent variable studies that favor domain-specific models.

38

Friday (May 22), 14:3015:30

We explored differentiation in WMC in two studies, applying a novel statistical

method, moderated factor analysis. Differentiation refers to the phenomenon that the
magnitude of correlations between tasks that tap different domains are inversely
related to ability. The first study (N= 5316) demonstrated the existence of
differentiaton in three complex span tasks: as capacity itself increases in a
population, WMC becomes domain-specific. The second study (N=249)
demonstrated that differentiation also occurs in WMC as the function of ability level in
fluid reasoning (Gf).
These results put the entire domain-generality of WMC debate in a new perspective,
and highlight the importance of the capacity level of the samples studied.
Key words: working memory capacity, differentiation, moderated factor analysis,
individual differences
Author contact e-mail address: kristof340@gmail.com

"But I was there!" - episodic memory as a mechanism for justification

Johannes Mahr, Central European University, Budapest, Hungary
Reka Peto, Etvs Lornd University, Budapest, Hungary
Gergely Csibra, Central European University, Budapest, Hungary

It is arguably a defining feature of episodic remembering that it provides one with

evidence that one had first-hand experience of a particular past event. So far there is
no agreed upon account of what the function of a memory system with such in-built
meta-representational features might be. We contend that a crucial function of
episodic memory is to make available reasons to justify one's claims about the past
through the use of source information. On this view, episodic memory is part of a
larger array of systems serving productive communication by ensuring that
assertions are justifiable. As such, episodic memory would ensure the veridicality of
beliefs about the past only in so far as it would be important in communicative
situations. Here, we present an experimental paradigm designed to test this
hypothesis.
Key words: episodic memory, source memory, meta-representation, justification
Author contact e-mail address: johannes.mahr@yahoo.de

39

Friday (May 22), 14:3015:30

Oxytocin modulates social categorization processes

Katalin Olh, Etvs Lornd University, Budapest, Hungary and Hungarian Academy of
Sciences, Budapest, Hungary
Jzsef Topl, Hungarian Academy of Sciences, Budapest, Hungary

Research from the past years has pointed out the importance of the neuropeptide
oxytocin in regulating various aspects of social interactions, such as inter-group
behavior. This study explores the question whether such effects may also be present
in social categorization processes. 24 adult males participated in the study with an
age range of 18-35 years. In the first part of the experiment, half of the participants
received intranasal oxytocin, while the other half was given placebo. After a 40minute-long waiting period, participants' categorization processes were assessed
with the memory confusion paradigm. Participants watched a presentation that
showed pictures of six adult men - three belonging to the Caucasian and three to the
African-American racial group. Each face was presented seven times and the
photographs were always accompanied with an utterance. In the test phase,
participants had to match the sentences to the faces from memory and the pattern of
errors was analyzed. The classical results obtained with this paradigm show that
people commit more within-group than between-group errors, which shows that
people organize information based on social category membership. Our preliminary
results indicate that this effect can be mitigated by the oxytocin administration.
Key words: oxytocin, social categorization, race, memory confusion paradigm
Author contact e-mail address: olah.katalin@ppk.elte.hu

Pragmatic aspects of reasoning in real life situations

Jelena Kuva Kraljevi, University of Zagreb, Croatia
Gordana Hrica, University of Zagreb, Croatia
Marijan Palmovi, University of Zagreb, Croatia
Marina Oluji, University of Zagreb, Croatia
Nevena Padovan, University of Zagreb, Croatia

It is a well established fact that people, when reasoning, do not follow the laws of
logic but provide information that they consider relevant for the discourse situation. In
fact, the inferences that are found in spoken language are often logically invalid.
In this study we analyze the Croatian corpus of spontaneous spoken language to find
patterns of reasoning and analyze them. In order to convey the intended meaning
participants in the conversation rely on the discourse markers. We concentrate on 'If then', 'therefore', 'moreover' and 'although' discourse markers as the facilitators of
logical or causal inferences.

40

Friday (May 22), 14:3015:30

The analysis will include two aspects of reasoning: logical form of the inferences
found in spoken language and the quantity and the precision of the information that is
exchanged in the discourse situation (e. g. how precise speakers are in telling time
regarding the discourse situation).
The corpus consists of 51 transcripts of spontaneous conversations between
Croatian adult speakers (20+) and is analyzed with CHAT/CLAN programme
package.
The study is conducted within the project "Adult language processing" (UIP-11-2013)
founded by Croatian Science Foundation.
Key words: reasoning, relevance theory, discourse, spoken language corpora
Author contact e-mail address: marina.olujic@erf.hr

Behavioral task as tool for short-term verbal memory and cognitive

vulnerability to depression assessment in mesial temporal lobe epilepsy
Lidija Preglej, Accredited Private Classical High School, Zagreb, Croatia
Draen Domijan, Department of Psychology, University of Rijeka, Rijeka, Croatia
Daniel Crnkovi, Department of Psychiatry, University Hospital, Zagreb, Croatia
Ksenija Marinkovi, Department of Psychology, San Diego State University, San Diego, USA
Hrvoje Heimovi, Department of Neurology, University Hospital, Zagreb, Croatia

Main generators of epileptic seizures in mesial temporal lobe epilepsy (MTLE) are
the amygdala and hippocampus, which are also associated with memory and
depression. We applied the free recall memory task in order to evaluate short-term
verbal memory and to explore possible cognitive vulnerability to depression in MTLE.
We also investigated the correlations between self-reported inventory scores and
memory results in MTLE subjects. Inventory scores included the Quality of life in
epilepsy inventory total score and its cognitive functioning domain score and BDI II.
We hypothesised that short-term memory in MTLE is impaired and cognitive
vulnerability to depression may be present. MTLE subjects (N=39) were paired with
healthy controls according to age, sex and education.
Compared to controls, MTLE subjects have significant memory deficit, measured by
number of correctly recalled words. Only in control subjects, memory for negative
valence words is significantly reduced in comparison with neutral or positive. Within
the MTLE group, memory is significantly reduced in its total score and for positive
valence words if depressive symptoms are present. The memory results significantly
correlate with each self-reported questionnaire score.
Key words: temporal lobe epilepsy, behavioural task, short-term verbal memory,
cognitive vulnerability to depression, self-reported subjective inventory
Author contact e-mail address: lidija_preglej@yahoo.com

41

Friday (May 22), 14:3015:30

Metamemory in educational psychology

Marta Riess, Jagiellonian University

The objective of this poster is to indicate important gaps in empirical research in the
field of educational psychology on the subject of metamemory. First of all the concept
of metamemory is rarely discerned from metacognition. Moreover metamemory is
traditionally believed to consist of two aspects: one is related to self-consciousness
and the other is the knowledge about the memory in general. One may be surprised
to find so little research has been done on the latter aspect, with regard to
interpersonal relations. This shortage is clearly problematic as far as understanding
of the process of teaching is concerned. Last but not least as the scarce research
concerning the interpersonal aspect of metamemory focuses solely on the
metacognitive capabilities of students, it neglects the question of the teachers'
capabilities. Filling this absence could lead to a better understanding of how
metamemory functions between people, and - ultimately - would help to improve the
teachers' training.
Key words: metamemory, educational psychology, review
Author contact e-mail address: riessmarta@gmail.com

The role of executive functions in patients with aphasia

Izabella Szllsi, Etvs Lorand University - Doctoral School of Education; National Institute
of Medical Rehabilitation, Hungary
gnes Lukcs, Budapest University of Technology and Economics - Faculty of Natural
Sciences - Department of Cognitive Science, Hungary

Recent studies suggest that together with language abilities executive functions
(EFs) can also be impaired in aphasia. Moreover, these functions can play a role in
language processing. We examined the updating of working memory (WM), and two
inhibition functions: prepotent response inhibition (PRI) and representational conflict
resolution in eight patients with aphasia with a lesion in their left frontal lobe, and
eight controls. Then we analyzed the relationship between EFs and language
performance.
We expected patients with aphasia to demonstrate lower performance on tasks
involving EFs than controls, and we also hypothesized that language comprehension
is going to be associated with updating and/or inhibition abilities in aphasia.
Our results showed that PRI and updating WM representations are both impaired in
aphasia. Performance on comprehension of grammatical structures and updating

42

Friday (May 22), 14:3015:30

WM representations correlated at a trend level, but we did not find relationship

between language comprehension and any of the inhibitory functions.
Taken together, generally observed linguistic symptoms of aphasia seem to be
accompanied by impairments of EFs (PRI and updating processes). The deficit in
updating might affect language comprehension in aphasia.
Key words: executive functions, aphasia, inhibition, updating
Author contact e-mail address: szollosi.iza@gmail.com

A comparison of possession comprehension in children with typical language

development and children with specific language impairments using E-prime
Ivona Radi Tatar,Polyclinic for the Rehabilitation of Listening and Speech SUVAG, Croatia
Paulina Mataija,Polyclinic for the Rehabilitation of Listening and Speech SUVAG, Croatia
Martina Vukovi Ogrizek, Polyclinic for the Rehabilitation of Listening and Speech SUVAG,
Croatia
Rafaela Marti, Polyclinic for the Rehabilitation of Listening and Speech SUVAG, Croatia
Marina Paprika, Polyclinic for the Rehabilitation of Listening and Speech SUVAG, Croatia

In Croatian, possession can be expressed in many ways, but the usage of

possessive adjective is prefered for animate possessor, and for inanimate possessor
prefered is the usage of preposition "od" followed by genitive "od+gen". The aim of
this paper was to compare and identify if there were differences in comprehension
and differentiation of correct language forms for expressing possession of animate
and inanimate possessor, and also in the speed of correct form detection. Two
groups of participants were included: 10 children with typical language development
(TLD) and 10 children with specific language impairments (SLI), aged 9-11, included
in speech and language therapy in Polyclinic SUVAG. The experiment was run using
the E-prime experimental software. Participants had to press a button if the
language form presented visually was correct. The results were surprising because
SLI children were more successful in detecting the correct form for expressing
possession of animate possessor than TLD children, while TLD children were more
successful in detecting correct form for expressing possession of inanimate
possessor. Results also showed that SLI children had longer reaction time.
Key words: possession, specific language imapairments, e-prime
Author contact e-mail address: iradic25@gmail.com

43

Friday (May 22), 14:3015:30

Deductive problem solving and learning logic

Mariusz Urbaski, Adam Mickiewicz University, Poland
Katarzyna Paluszkiewicz, Adam Mickiewicz University, Poland
Joanna Urbaska, Adam Mickiewicz University, Poland

Is there any correlation between fluid intelligence (Gf) and fluencies in both easy
and difficult deductions? Does learning logic have any impact on the level of
fluencies in both kinds of deductions?
Tools: Raven's APM (alpha = .83), Deductive Reasoning Test (DR, alpha = .64),
Erotetic Reasoning Test (ER, alpha = .78), Polisyllogisms Test (PS; McDonalds
omega total = .8); DR, ER and PS were designed by the authors, operationalizing
easy (DR) and difficult (ER and PS) deductions.
Subjects: Our subjects (154 students, M=21,69, SD=1,44) formed groups of lower
(group C) and higher (groups A and B) abilities with respect to fluid intelligence and
of extensive (group A) and very limited (groups B and C) training in formal logic.
Results: Groups A and B performed better than group C in APM, DR and ER. Group
A obtained significantly higher results than group B in ER and PS; their performance
in APM and DR were comparable. We conclude on this basis that deductions of
different complexities call for different abilities to be manifested and that fluency in
difficult deductions, while related to Gf, depends also on subjects experience and
that this does not hold in case of simple deductions.
Author contact e-mail address: murbansk@amu.edu.pl

The role of inferential direction on psychotherapists metacognitive confidence

and need for more information
Jacinto, Sofia, Universidade de Lisboa; Indiana University
Ferreira, Mrio, Universidade de Lisboa;
Braga, Joo, Universidade de Lisboa;

The non-decompos able nature of a psychotherapy sessions favors the use of

intuition (Hammond et al., 1987). The feeling of rightness associated to intuitive
judgments (e.g., Koriat, 2012, Thompson et al., 2012) may lead to therapists
overconfidence and to a decreased need for seeking further information. This may
compromise therapists willingness to critically evaluate their initial clinical (intuitive)
judgments.
We hypothesize that this tendency could be moderated if besides causal
explanations (backward inferences) therapists were requested to make predictions
(forward inferences) based on the same session information. Predictions are

44

Friday (May 22), 14:3015:30

inherently associated to more uncertainty than causal explanations (Hogarth, 2010)

therefore they could decrease overconfidence in the clinical judgment and promote
further information seeking.
To test for this, psychology students were asked to find causal explanations and
make predictions about a hypothetical client. Preliminary results showed, as
expected, higher uncertainty associated with predictions. Moreover, participants
tended to use their backward inferences instead of their predictions to prepare their
future clinical work. Strategies to avoid the overconfidence stemming from first
intuitions are discussed.
Author contact e-mail address: anajacinto@campus.ul.pt, sofiabjacinto@gmail.com

45

Saturday (May 23), 10:3011:30

Session 3
Solvable or Not Solvable? Heuristic Cues that Underlie the Judgment of
Solvability
Yael Beller, Technion--Israel Institute of Technology, Israel
Rakefet Ackerman, Technion--Israel Institute of Technology, Israel

Meta-reasoning is an emerging domain within the Metacognitive research, which

deals with reasoning and problem solving. Judgment of Solvability (JOS) reflects the
assessed probability that a problem is solvable. Initial JOS takes place upon
approaching a problem and final JOS is done after a solving failure, and may be a
basis for deciding whether to try again. To date, the bases for these judgments are
unknown. Response latency and accessibility (amount of associations that come to
mind regarding a question) are well-established heuristic cues for meta-memory
judgments. Our hypothesis was that these cues are predictive also of both JOS
types, but not necessarily in the same manner as in meta-memory. Solvable and
unsolvable compound remote associates (CRA) were used. In Experiment 1 we
examined response latency and accessibility as cues for JOS. As predicted, quick
response latency predicted high initial JOS, as in meta-memory. However, the
reverse pattern was found for final JOS. High accessibility predicted high JOSs, but
this was a misleading cue, as in fact success rates dropped with accessibility. In
Experiment 2 we examined the same cues when judging memorizing of the same
stimuli. The comparison revealed shared and distinct mechanisms of meta-reasoning
and meta-memory.
Author contact e-mail address: ackerman@ie.technion.ac.il

Assessment of Non-Native Speakers' Metacognitive Abilities and Reasoning

Stem from Rational Thinking to Adopt a New Template of Language
Grammatical Aspects: Evidence from MEG and fMRI
Sadeq Ali Saad Al Yaari, University of Zagreb

Purpose: Using MEG and fMRI, the aim of this study was to investigate non-native
speakers' metacognitive abilities and reasoning stem from rational thinking to adopt a
new template of Arabic grammatical aspects.
Design and Method: In this quantitative and qualitative study, two measurement tools
were used to evaluate participants' performance: Two tests of grammatical
production and neuroimaging test.

46

Saturday (May 23), 10:3011:30

Settings: Al Khars hospital, KSA.

Participants: Twenty-three to thirty-year-old participants from 11 different nationalities
(N = 14 all males).
Procedures: A within-groups design and a between-groups design are used to elicit
participants' performance before comparing it in light of MEG and fMRI outcomes.
Interventions: Data were neurolinguistically analyzed before it was statistically
visualized.
Main outcomes and Results: Non-native speakers' un-exposing to foreign/second
languages in their critical period age disable them to adopt a new grammatical
system easily. MEG can identify language lateralization and diagnose silent reading,
while fMRI helps distinguish declarative knowledge from procedural knowledge.
Conclusions & Implications: Non-native speakers' grammatical mistakes could be
avoided if we understand decoding processes in both audioceptive and
ophthalmoceptive centers in the brain. Through signals, MEG and fMRI found to be
effective means through which sensory cortex activation and language lateralization
and localization can be differentiated.
Author contact e-mail address: salyaari@ksu.edu.sa

Overcoming intuitive interference in geometry: from brain studies to an

educational intervention
Reuven Babai, Tel Aviv University, Israel
Enav Shalev, Tel Aviv University, Israel
Ruth Stavy, Tel Aviv University, Israel

Students difficulties in mathematics and science may stem from interference of the
tasks salient irrelevant variables. Here, we focus on a comparison of perimeters
task, in which the area is the irrelevant salient variable. In congruent trials (no
interference), accuracy is higher and reaction time is shorter than in incongruent trials
(area variable interference).
A brain-imaging study indicated that correctly answering the incongruent condition is
associated with activation in prefrontal brain regions known for their executive
inhibitory control. These findings suggested that intervention aimed at activating
inhibitory control mechanisms could improve students success.
We explored the effect of an intervention that explicitly warns about the possible
interference of the variable area. Eighty-four sixth graders performed the same test,
with warning intervention (warning group) or without it (control group).
Accuracy in the warning group was significantly higher in incongruent conditions and
reaction time was significantly longer in all conditions than in the control group. The
results suggest that the explicit warning activates inhibitory control mechanisms and
thus helps students overcome the interference.

47

Saturday (May 23), 10:3011:30

Our findings point to the possibility of improving students problem-solving abilities

through simple and focused interventions that explicitly warn them about the trap in
the task.
Author contact e-mail address: reuvenb@post.tau.ac.il

Contaminated mindware and reflective mind

Vladimra avojov, Rbert Hank, Eva Ballov Mikukov,
Institute of Experimental Psychology, Slovak Academy of Sciences, Slovakia

Our study was inspired by Stanovichs (2011) proposed framework of assessing

rational thinking. We used most of the tasks hypothesized to tap into resistance to
miserly processing (e.g. resistance to belief bias, outcome bias, denominator neglect,
cognitive reflection, disjunctive reasoning, and diagnostic hypothesis testing) and
tried to verify whether some irrational beliefs (the contaminated mindware - CM) or
thinking dispositions can serve as inhibitors of rational thinking.
596 students (mean age=19.8 years) completed battery of measures consisting of
intelligence test, thinking dispositions measures, irrational beliefs, and rational
thinking tasks. Only thinking dispositions proved to be predictors of whether a person
will hold irrational beliefs, but both preferences for analytical as well as intuitive
thinking styles predicted a higher degree of irrational beliefs. To test the hypothesis
that CM inhibits rational thinking, we performed moderation analysis, where cognitive
ability was a predictor, CM was the moderator, and the score in cognitive reflection
test was an output. Model parameters were: R2=.077; p=.00004; n=287; R2 increase
due to interaction R2=.00007; p=.883. We found no support for the claim that CM
inhibits cognitive reflection and we discuss the results in the paper.
This work is part of research project Centre of Excellence SAS 'Centre for Strategic
Analyses "(CESTA) III / 2 /, 2011.
Key words: cognitive reflection, irrational beliefs, thinking dispositons, intelligence
Author contact e-mail address: vladimira.cavojova@savba.sk

48

Saturday (May 23), 10:3011:30

The impact of social context on children's executive functioning and it's

relationship to creativity
Paula Fischer, University of Edinburgh
Nicolas Chevalier, University of Edinburgh
Mark A. Runco, University of Georgia

Executive Functioning (EF) refers to a problem-solving process that develops during

childhood, with a shift from mainly reactive control to flexible engagement of reactive
and proactive control with age. EF, similar to other cognitive functions, develops
within social context. Children profit from being with an adult, an older child, or a peer
in a variety of social contexts. Cooperation has been hypothesized to enhance goal
maintenance which can lead to proactive behaviour, and competition has not yet
been tested among pre-schoolers. In the current study, we examined whether a
social context such as cooperation or competition would enhance pre-school
children&#8217;s cognitive control strategy from reactive to proactive control, using
an AX-CPT continuous performance task. Furthermore, we implicated a 3D Alternate
Uses Task to test divergent thinking, as a component of creativity. Our results
indicated significant effects of condition. Both cooperation and competition enhanced
children's executive performance, although they continued to use a reactive control
strategy. We also found that children who performed more accurately in the
executive functioning task showed a poorer performance in divergent thinking,
suggesting that the lack of appropriate cognitive control strategy may be actually
beneficial at this age in certain circumstances.
Key words: executive functions, social context, cooperation, creativity
Author contact e-mail address: paula.fischer1989@gmail.com

14-month old infants represent others' beliefs about the number and identity of
objects
Dora Kampis, Central European University, Hungary
gnes Melinda Kovcs, Central European University, Hungary
:

Infants understanding of others beliefs has been investigated through a wide range
of tasksWe tested the proposed limitations of infants mindreading capacities that
should not extend to beliefs regarding object identity or numerosity (Butterfill &
Apperly, 2013). We used a manual search paradigm where infants search longer if
they think there is still an object present (Feigenson & Carey, 2003). Infants saw a
scene where 1 (in Study 1, 3 and 4) or 2 (in Study 2) objects were put into a box by
Experimenter 1 (E1). Then a further object was added (Study 1), one was taken out
(Study 2), exchanged to another object (Study 3), or transformed into another

49

Saturday (May 23), 10:3011:30

appearance (Study 4) by Experimenter 2 (E2). During this E1 could be present (True

Belief condition) or absent (False Belief condition). Finally, E1 took out one object
from the box. We measured how long infants searched for an object. Results show
that search times were influenced by the belief of E1 regarding the number of objects
[0/1] remaining in the box. This suggests that infants successfully tracked the other
persons belief when it involved multiple objects; while taking into account the identity
of objects involved.
Author contact e-mail address:

Metacognition and performance rating in spatial manipulation tasks

Roland Kasek, KRE Institute of Psychology, Hungary
Szabolcs Takcs PhD, KRE Institute of Psychology, Hungary
Mnika Albu PhD, KRE Institute of Psychology, Hungary

Aim: The aim of the study is to reveal how metacogniton promotes performance in
basic spatial cognition tasks.
Method: The study examines the estimated performance of subjects grouped by
profession (designers and non-designers) in four spatial manipulation tasks of
increasing complexity (Kasek & Albu, 2014). The baseline task (Task1) is a modified
Corsi where homogenous stimuli (dots) are displayed sequentially and requires
correct localization. Task2 requires spatial recall of simultaneously displayed
numbers in numeric order. Task3 repeats Task2 in self-paced timing (measured by
inspection time; index A of metacognition). Task4 requires numeric interferenceinhibition and correct localization. After each task, subjects are asked to rate their
performance in comparison with their virtual peers (index B of metacognition).
Results: Results show positive metacognition effect on localization in Task3 by index
A. Subjects' estimations of their own and their virtual peers performance - measured
by index B - were precise. However, based on the results, designers tend to
somehow underestimate their own performance.
Conclusions: Metacognition plays an important role in basic spatial cognition,
includes monitoring and estimation of both self and others performance. On the other
hand, it seems that type of expertness (e.g. designers) can moderately affect
metacognition effectiveness.
Key words: metacognition, performance rating, spatial cognition, expertness
Author contact e-mail address: roland@kasek.com

50

Saturday (May 23), 10:3011:30

Negative numbers in elementary school

Kiss Orsolya, Budapest University of Technology and Economics, Hungary
Krajcsi Attila, Etvs Lornd University, Hungary

In our research we assume that the understanding of the numbers derives from the
system in which they are anchored. Our hypothesis is that children discover the
meanings of number words based on their knowledge about the objects, number
words being anchored in the understanding of the objects semantic system, and
matching sets. Children are able to perform those numerical operations that can be
performed on the objects as well. Furthermore we assume that numbers could be
anchored to other semantic systems too. Using a different kind of anchoring we can
expand the set of meaningful operations, because they can be transferred from the
newly anchored semantic system to the numbers.
We test this model with the learning of negative numbers. We expect that in the case
of the original object based analogy the negative numbers are hard to understand.
However, anchoring the numbers to a spatial number line can help understanding the
operations on negative numbers.
Participants included first and second graders (8 and 9 years olds). In order to test
the two possible anchorings, we gave the children simple arithmetical operations
such as addition and subtraction, using different counting tools: either number line or
marbles. According to the results children make less error with number line than with
marbles. This result is in a dissociation with our former result showing that younger
children prefer objects over a number line for operations with natural numbers. We
conclude that anchoring the abstract layer of numbers to several domains can extend
the capabilities of numerical understanding. This multiple anchoring is a key
component in abstract mathematical thinking and flexible numerical processing.
Key words: negative numbers, semantic system, anchoring
Author contact e-mail address: kisorsolia@gmail.com

Low cognitive load and reduced arousal impede practice effects on executive
functioning, metacognitive confidence and decision making
Sabina Kleitman, University of Sydney, Australia
Simon A Jackson, University of Sydney, Australia
Eugene Aidman, Defence Science and Technology Organisation, Australia

We investigated the effects of low cognitive workload and the absence of arousal
induced via external physical stimulation (motion) on practice-related improvements
in executive (inhibitory) control, short-term memory, metacognitive monitoring and

51

Saturday (May 23), 10:3011:30

decision making. Office workers (N=70) performed low and moderately engaging
tasks in two successive 20-minute simulated drives and repeated a battery of
decision making and inhibitory control tests three times before, between and after
drives. For half the participants, visual simulation was synchronised with (moderately
arousing) motion. Others performed the same drive without motion. Performance
significantly improved over the three test blocks, indicative of practice effects.
Improvement was highest when both motion and moderate cognitive load were
present. The same effects declined in the absence of motion or following a low
cognitive workload task, suggesting two distinct pathways through which practicerelated improvements in cognitive performance may be hampered. Practice,
however, degraded certain aspects of metacognitive performance, as participants
became less likely to detect incorrect decisions in the decision-making test with each
subsequent test block. Implications include consideration of low cognitive load and
arousal as factors responsible for performance decline and targets for the
development of interventions/strategies in low load/arousal conditions such as
autonomous vehicle operations and highway driving.
The authors were funded by the Defence Capability Development Group Project
Land 121 Phase 4 (http://www.army.gov.au/Our-future/Projects/Project-LAND-121)
under the Research Agreement 558023.
Author contact e-mail address: sabina.kleitman@sydney.edu.au

The Source of the Distance Effect in an Artificial Number Sequence

Petia S. Kojouharova, Etvs Lornd University, Hungary
Attila Krajcsi, Etvs Lornd University, Hungary

Distance effect in numerical cognition means that the closer two quantities are
numerically, the more difficult it is to differentiate between them. However, instead of
originating from the numerical meaning (i.e., values) of the numbers as in 5-2=3, the
distance effect might reflect a surface characteristic such as the connection of the
labels.
While the meaning and the label connections correlate in normal circumstances (e.g.,
in the Indo-Arabic notation), artificial number sequences with gaps (e.g., 1, 2, 5, 6)
can dissociate the two properties. In a sequence with gaps the meaning model
predicts that the distance effect changes with the values of the digits (e.g., the 2-5
distance is 3), while the label model predicts that the distance effect changes with the
position of the digit in the incomplete series (e.g., the 2-5 distance is 1).
Participants learned artificial symbols for the 1, 2, 3, 7, 8, and 9 numbers and then
compared them in pairs in which they had to choose the numerically larger number.

52

Saturday (May 23), 10:3011:30

Both models were able to explain our data to some extent, thus suggesting that both
label and meaning jointly contribute to the distance effect.
Key words: numerical cognition, distance effect
Author contact e-mail address: kojouharova.petia@ppk.elte.hu

The Trusting Twos: The epistemic weight attributed to communication

increases during toddlerhood.
Mascaro Olivier, Central European University, Hungary;
Agnes Kovacs, Central European University, Hungary

We report data suggesting that the epistemic weight given to communicated

information varies during development. Infants appear to be surprisingly reliant on
communication, even when it conflicts with previous perception. Moreover, contrary
to the view that children become more skeptical with age, we find that trust in
communication increases during toddlerhood.
In Study 1, 15-month-olds have to find a toy hidden under one of two buckets. They
first see where the toy is hidden, and later an informant tells them that the toy is in
the other bucket. Contrary to what adults would do, children trust communicated
information rather than their past perception, even when they have evidence that the
communicators belief about the toys location is false. Controls indicate that children
do not go along with what the informant tells them just to please her, and rather
genuinely believe her.
Study 2 shows that humans trust in communication increases during the second
year of life. In a setting identical to Study 1, twenty-four-month-olds trust
communication even more than 15-month-olds.
Increased communicative abilities and opportunities to learn from others could justify
the Trusting Twos, a developmental stage of heightened trust in communicated
information.
Key words: trust, information integration, naive epistemology, communication
Author contact e-mail address: olivier.mascaro@gmail.com

Is writing metarepresentational?
Olivier Morin, KLI Institute, Klosterneuburg, Austria

The standard view of writing, dating back to Aristotle, sees writing as a derivative
code: a secondary representation of a natural language. This view has been
challenged. It has been noted that writing cannot be said to represent speech, but

53

Saturday (May 23), 10:3011:30

rather, more abstract properties of natural language.s The status of visual codes
such as pictographies, mathematical or musical notations, etc. can also be discussed
in this connexion. This poster will defend a revamped version of the standard,
aristotelian thesis. Writing will be defined as a general-purpose visual code for
asynchronous communication. It is distinct from both sign language (which is
attached to synchronous communication) and specialised graphic codes. I will offer a
conjecture as to why general-purpose codes for asynchronous communication tend
to be backed by a natural language.
Key words: writing, language, semiotics
Author contact e-mail address: olivier.morin@kli.ac.at

Affect and Metacognition: Confidence (in)Sensitivity When Feeling Happy

Yael Sidi, The Technion Israel Institute of Technology, Israel
Rakefet Ackerman, The Technion Israel Institute of Technology, Israel
Amir Erez, University of Florida, Florida

The relationship between positive affect and metacognition has been scarcely
studied. The present research examined the influence of induced positive affect on
metacognitive processes while answering challenging general knowledge questions.
Participants were induced with positive or neutral affect by a picture-tagging task.
Based on previous findings, we expected positive affect to facilitate memory search
and boost confidence compared to a neutral affective state. Our aim was to examine
whether this confidence boost would reliably reflect task-related performance
fluctuations. Two manipulations took place for influencing performance without
effecting actual knowledge: answering format and social motivation. We
hypothesized that positive affect will generate insensitivity of confidence ratings to
task-related performance fluctuations. All predictions were supported. In particular,
positive affect did enhance performance and confidence relative to the neutral affect.
However, compared to the neutral affective state, under positive affect participants
exhibited larger overconfidence and were less sensitive to performance fluctuations.
Our results suggest that although positive affect facilitates cognitive performance, it
impairs metacognitive monitoring. Practical and theoretical implications for both affect
and metacognition bodies of literature are discussed.
Key words: positive affect, metacognitive monitoring, general knowledge,
overconfidence
Author contact e-mail address: yaelsidi@tx.technion.ac.il

54

Saturday (May 23), 10:3011:30

The Roles of Intention and Belief on Mood

Sabina Skubic, University of Ljubljana, Slovenia
Alexander Batthyany, University of Vienna, Austria

My empirical study was set to explore whether eating apples blessed by Hare
Krishna would enhance mood more than eating ordinary apples and how belief on
which apples you receive influence your mood. 20 participants were randomly
assigned into two groups, one receiving blessed and one non-blessed apples, which
was not revealed until after the experiment. They filled out a questionnaire before
eating apples which measured sensory processing sensitivity, an impact of their
mindset on their mood, religious and paranormal beliefs, their mood in that moment
and general likeness of apples. After eating given apples they indicated which apples
they thought they received and filled out the PANAS questionnaire, measuring their
mood just after receiving apples.
Analysis of the results is still in the process. Participants are to be divided into four
groups, depending on which kind of apples they received and on their beliefs of
which apples they have received. My hypothesis is that apples treated with good
intention improve mood more than ordinary apples and that belief and intention
interact, meaning that belief that one is eating a blessed apple improves mood more
if the apple is truly blessed.
Key words: intentional food, mind-set, belief, mood
Author contact e-mail address: sabina.skubic@gmail.com

Metacognition at a collective level

Tobias Strzinger, University of Stuttgart

Metacognition, understood as the epistemic self-evaluation of cognitive processes,

enables humans to prospectively predict the outcome of mental actions and
retrospectively decide whether a performed mental action was successful. Since
many philosophers have argued that cognition can also be identified at a level above
the individual, namely at a distributed or collective level, it is worth asking whether
the possibility of being distributed or collective also applies to metacognition. I.e., are
there any forms of collective or distributed epistemic self-evaluations that let
collectives or distributed systems somehow control and monitor their cognitive
processes? List and Pettit have argued that group reasoning is only possible within
an 'informal organizational structure' and that explicit and formal organizational
structures can only establish e.g., consistency by aggregating individual beliefs
through a special way of voting. Contra List and Pettit I

55

Saturday (May 23), 10:3011:30

argue that a (metacognitive) deliberative process can also be a vital part of explicit
organizational structures because the process of generating individual beliefs, upon
which the collective beliefs supervene, is not an individualistic and isolated truth
tracking mechanism but a collective exchange of arguments. The metacognitive
function of a collective is then realized through deliberation before voting to generate
a group belief.
Key words: Collective Metacognition, Distributed (Meta-)cognition
Author contact e-mail address: Tobias.Stoerzinger@philo.uni-stuttgart.de

On the cognitive biases as cognitive territories outside positive manifold

continent
Predrag Teovanovi, University of Belgrade, Republic of Serbia
Lazar Stankov, Australian Catholic University, Australia

Study was aimed to address questions of generality of individual differences in

cognitive biases and their reducibility to other well-known cognitive measures.
Heterogeneous set of seven cognitive biases was selected (belief bias, anchoring
effect, overconfidence bias, hindsight bias, outcome bias, base rate neglect, and
sunk cost effect) and tests for the measurement of individual differences in these
biases were designed. They were computer administered to participants (N=243)
together with Raven's Matrices, Swaps, 3D Space, Vocabulary, Analogies,
Synonyms-Antonyms, Need For Cognition, Intellect/Openness, and Cognitive
Reflection Test.
Mean bias scores deviated from normative values by between 1.64 and 3.58
standard deviations, indicating large effects of normatively irrelevant variables.
Cronbach's alphas were above .70 for all measures but hindsight bias, thus
suggesting that individual differences in susceptibility to cognitive biases can be
reliably measured. However, their generality was very low. Correlations among bias
measures were small (r's<.20) and both positive and negative Also, only some of
them were related to measures of intelligence and CRT (r's<.25), none was related to
NFC and O/I. In sum, results indicate that main part of reliable variance in
susceptibility to cognitive biases is unique and support the conclusion that processes
captured by cognitive biases are diverse.
Author contact e-mail address: teovanovic@gmail.com

56

Saturday (May 23), 10:3011:30

The development of metacognitive monitoring-control processes in adolscence

Nike Tsalas, Ludwig Maximilian University of Munich, Germany
Markus Paulus, Ludwig Maximilian University of Munich, Germany
Beate Sodian, Ludwig Maximilian University of Munich, Germany

The interplay between metacognitive monitoring and control improves during

childhood, such that 9- but not 6-year old children allocate more study time to difficult
than to easy learning pairs. The current study explored whether such self-regulation
continues to develop in adolescence and whether participants perform differently
under self-paced and timed conditions. We tested a group of 10- (N=31) and 14-year
olds (N=29) and adults (N=29) in a within-subjects design. In each of two consecutive
blocks, participants first monitored their learning of easy and difficult picture pairs
through Judgments of Learning (JoL) and then studied the first set of pairs in a selfpaced manner and the second set of pairs under time pressure. Results showed that
in both conditions all age groups allocated more study time to items with low than
with high JoL. The return-rate of learning (proportion of items correct/time invested)
showed that adults studied in a more efficient manner compared
to 10- and 14-year olds. Moreover there was a higher learning rate in the timed,
compared to the self-paced condition for all age groups. These results suggest that
efficient study time allocation continues to develop in adolescence and that even
adults labour in vain when time is abundant.
Research funding from the National Research Fund Luxembourg and partly from a
European Research Council (ERC) advanced grant (Project 269616).
Author contact e-mail address: Nike.Tsalas@psy.lmu.de

When problem solving meets metacognition: thinking about writing before

doing it.
Rosa Volpe, Universit Nice Sophia Antipolis, France
Lucile Chanquoy, Universit Nice Sophia Antipolis, France

Introduction:
Writing is a complex task requiring the triggering of problem solving abilities
(Alamargot & Chanquoy, 2001). The role metacognition plays on the written
processes has met little attention within this field of research, with few exceptions.
Goal of the experiment: This study addresses the following questions : can primary
school children develop metacognitive knowledge about writing? It predicts that
encouraged to develop metacognitive knowledge about writing, young children
should manage to integrate such knowledge into their productions.
Method:

57

Saturday (May 23), 10:3011:30

Subjects: Third, Fourth and Fifth graders participated to this study and were divided
into experimental and control group.
Material: 4 metacognitive questionnaires on the process of writing for the
experimental group. 4 questionnaires about the writing process for both groups.
Procedure: Before answering to the open questions about writing experimental group
children had to mark their agreement/disagreement with the statements within each
metacognitive questionnaire which aimed at developing their metacognitive
awareness and knowledge about writing.
Results:
Results show that the children within the experimental group became metacognitively
aware, and they integrated such metacognitive knowledge into their answers to the
questions. Compared to the control group, the experimental group integrated a
higher number of metacognitive items into their productions.
Author contact e-mail address: rosa.volpe@unice.fr

The Effect of Explicit and Implicit Category Learning on Classification

Accuracy and Confidence
Valnea auhar, Department of Psychology, Faculty of Humanities and Social Sciences,
Rijeka, Croatia;
Igor Bajanski, Department of Psychology, Faculty of Humanities and Social Sciences,
Rijeka, Croatia;
Draen Domijan, Department of Psychology, Faculty of Humanities and Social Sciences,
Rijeka, Croatia;

The aim of study was to investigate how activation of explicit and implicit
categorization systems during learning affects classification accuracy and confidence
judgments in the later test phase. One group of participants was informed about the
correct classification rule at the beginning of the learning phase (rule condition).
Another group was never informed about the rule (no-rule condition).
In the test phase, old and new items were presented. New items differed from the old
items with respect to transfer type (good or bad) and item consistency (consistent or
inconsistent). Good transfer items belonged to the same category as their old twin
item, while bad transfer items belonged to the opposite category. Consistent items
were highly similar to their old twins, while inconsistent items were less similar.
In the rule condition, the results showed that neither transfer type nor item
consistency affected classification accuracy or confidence judgments. In the no-rule
condition, the results showed that transfer type affected accuracy, but not
confidence. On the other hand, item consistency affected confidence but not
accuracy. The results suggest that explicit category learning system is activated in

58

Saturday (May 23), 10:3011:30

the rule condition, while implicit category learning system is activated in the non-rule
condition.
Key words: category learning systems, classification accuracy, confidence judgments
Author contact e-mail address: vzauhar@ffri.hr

59