KBKatarina Begus

Katarina Begus

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article

Previous Organizations / Institutes

The organizations / institutes from the scholar's other articles in the past ten years


Katarina Begus has a total of 32 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    Infants trade off reward for information: longitudinal evidence for curiosity-driven information-seeking in the second year of life

    by Cécile Gabrielle Gal, Dimitris Askitis, Katarina Begus

    Curiosity-driven information seeking is a crucial behaviour thought to drive our progress as a species. Valuing information as rewarding has been established as a driver of curiosity in human adults and non-human primates, characterised by a stringent set of three criteria formalized by work in primates, but direct evidence in human infants remains absent. We followed N=125 infants longitudinally over three visits (12, 18, 24 months) and used an infant-controlled eye-tracking paradigm to test whether valuing information as rewarding could also underlie curiosity in infancy. In each trial, infants could reveal two hidden characters by looking at their hiding locations, yielding either high or low information-gain. Revealing the first character was coupled with a rewarding animation, while revealing the second character required to disengage and thereby trade-off the ongoing animation for information. We found evidence in support of all three curiosity criteria: throughout their second year of life, infants were willing to forego extrinsic rewards to obtain information , even when the information was strategically useless, and – from 18 months – in a way that scaled with information gain. Infants did not meet the third criterion at the youngest age, which we attribute to age-related cognitive-load effects. Individual differences in infants' sensitivity to information gain showed modest stability across the visits. Overall, this study establishes reward-based information-seeking as a conserved mechanism present from early infancy, with early individual differences that may carry lasting consequences for learning and cognitive development.

  • Children’s exploration distinguishes necessary and merely possible outcomes.

    by Luisa Andreuccioli, Sophie Mazor, Katarina Begus, Elizabeth Bonawitz, Stephanie Denison, Caren M. Walker

  • Open Access

    Early Cognitive Offloading: Three-Year-Olds Preferentially Offload Goal-Relevant Information

    by Julieta Medawar Dumandzic, Dawa Dupont, Katarina Begus

    Cognitive offloading, the use of physical actions to reduce cognitive demands, is well studied in adults, but little is known about when this ability emerges in development. While older children show flexible, goal-directed offloading sensitive to memory load, younger children's offloading has typically appeared reactive and imitative rather than self-generated. Existing paradigms often present offloading as a strategy children must explicitly recognise as useful, measure it as a binary choice, and place its benefits at a later retrieval stage, thereby placing demands that may exceed young children's comprehension. We tested whether 3-year-olds (N = 46) could engage in selective cognitive offloading through a child-friendly, touchscreen-based task. Prior to a memory-dependent delivery phase, children could move baskets containing target or non-target items closer to their destinations, with the opportunity to offload time-limited such that not all items could be moved. Children preferentially offloaded target over non-target items, and both the effectiveness and selectivity of their offloading were positively associated with delivery accuracy. Offloading also significantly improved performance relative to an unaided memory condition, and this benefit extended even to items children had not moved during offloading. However, offloading behaviour did not increase under higher memory load, nor was it related to individual differences in unaided memory accuracy, unlike the demand-sensitive, compensatory patterns typical of older children and adults. These findings suggest that selective, goal-directed offloading emerges earlier than previously shown, but may initially reflect basic goal-directed action rather than deliberate metacognitive regulation that characterises offloading later in development.

  • Open Access

    I err, therefore I am: Neural signature of error-monitoring predicts self-awareness in infants

    by Cécile Gal, Dimitris Askitis, Katarina Begus

    Self-reflection is central to human cognition and arguably a defining feature of human consciousness, yet the mechanisms enabling our mind to reflect upon itself and to construct a sense of self are unclear. Specifically, whether metacognition implies self-awareness, is independent of it, or emerges because of it, remains a subject of debate, fuelled by reports of metacognitive-like behaviour in preverbal infants and non-human animals. Building on evidence that 12-month-old infants show neural signatures of internal error-monitoring, we follow infants longitudinally from 12 to 18 months, when infants first start exhibiting evidence of a conceptual self-representation. Using a gaze-contingent match-to-sample paradigm with concurrent eye-tracking and EEG, we confirm that infants exhibit a neural signature of internal error-monitoring (error-related negativity, ERN) after making an incorrect decision (saccadic eye-movement) and before receiving visual feedback. Critically, we find that ERN at 12 months developmentally predicts which of the infants will evidence self-representation (mirror self-recognition) six months later. Infants’ self-recognition was predicted specifically by internal error-monitoring, and not by external feedback-monitoring (FRN), nor by infants’ age or general development. Moreover, we show that only infants who later self-recognised adjusted their behaviour post-error, by slowing their decisions and increasing visual exploration. Thus, internal error-detection, if available, can be utilised for online adjustments in information-seeking and decision-making as early as 12 months of age. These converging neural and behavioural findings support the idea that, rather than presupposing a self, early metacognition may be functionally involved in building conceptual self-awareness in humans.

  • A Socio-Epistemic Theory of Attachment: Part I. Secure Attachment as Epistemic Trust and Its Place in Development

    by Alessandro Talia, Robbie Duschinsky, Diana Mazzarella, Katarina Begus, Johann R. Kleinbub, Madeleine Miller-Bottome, Svenja Taubner, Jeremy Holmes, Peter Fonagy

  • A Socio-Epistemic Theory of Attachment: Part III. A Relevance-Based Taxonomy for the Adult Attachment Interview (AAI) Coding System

    by Alessandro Talia, Robbie Duschinsky, Diana Mazzarella, Katarina Begus, Johann R. Kleinbub, Madeleine Miller-Bottome, Svenja Taubner, Jeremy Holmes, Peter Fonagy

  • A Socio-Epistemic Theory of Attachment: Part II. Adult Attachment Classifications as Differences in Fostering Epistemic Trust

    by Alessandro Talia, Robbie Duschinsky, Diana Mazzarella, Katarina Begus, Johann R. Kleinbub, Madeleine Miller-Bottome, Svenja Taubner, Jeremy Holmes, Peter Fonagy

  • Open Access

    A Socio-Epistemic Theory of Attachment: Part I. Secure Attachment as Epistemic Trust and Its Place in Development

    by Alessandro Talia, Robbie Duschinsky, Diana Mazzarella, Katarina Begus, Johann R. Kleinbub, Madeleine Miller-Bottome, Svenja Taubner, Jeremy Holmes, Peter Fonagy

    This article, written in honor of Mary Main’s legacy, introduces a new socio-epistemic theory ofattachment and attachment-related differences. Inspired by Main’s pioneering work on attention and discourse, we integrate insights from epistemic trust theory (Fonagy & Allison, 2014) and cognitive pragmatics (Sperber & Wilson, 1986) to argue that attachment patterns are not specifically adaptations to caregivers’ responsiveness to distress, but broader strategies for optimizing communication. Namely, we propose that attachment patterns reflect variations in epistemic trust towards the caregiver—the trust in the relevance and authenticity of interpersonal communication. Our framework helps explain key findings of attachment research, including the so-called 'intergenerational transmission of attachment,' as well as the association between attachment differences and a broad range of developmental outcomes seemingly unrelated to threats and safety, but to communication and learning. In Part I, we examine how infants adjust their epistemic trust in response to caregiver communication, shaping attachment patterns and later development. In Part II (forthcoming), we extend this model to a view of AAI classifications as differences in fostering epistemic trust. In Part III (forthcoming), we apply our perspective to explain AAI subclassifications, the unresolved AAI classification, mentalizing, and more, further celebrating Main’s legacy in attachment theory and research.

  • Open Access

    A Socio-Epistemic Theory of Attachment: Part II. Adult Attachment Classifications as Differences in Fostering Epistemic Trust

    by Alessandro Talia, Robbie Duschinsky, Diana Mazzarella, Katarina Begus, Johann R. Kleinbub, Madeleine Miller-Bottome, Svenja Taubner, Jeremy Holmes, Peter Fonagy

    This article builds upon the socio-epistemic theory of attachment introduced in Part I, proposing that adult attachment classifications, as measured by the Adult Attachment Interview (AAI), reflect differences in fostering epistemic trust. While the AAI has been shown to reliably predict caregiving sensitivity and infant attachment patterns, questions remain regarding the precise nature of what it measures. We argue that attachment-related differences in the AAI reflect generalized communication tendencies rather than merely responses to attachment-related content. Drawing on Relevance Theory, we suggest that secure, dismissing, and preoccupied AAI classifications correspond to distinct ways of optimizing relevance in communication— balancing informativeness and processing effort in ways that shape epistemic trust. We illustrate this framework through analyses of AAI discourse patterns and psychotherapy transcripts, demonstrating that these communicative tendencies extend beyond attachment-specific contexts. This perspective offers a unified account of attachment representation, communication styles, autobiographical narratives, and intergenerational transmission of attachment, reframing attachment-related discourse differences as variations in the ability to establish epistemic trust across relationships and contexts.

  • Open Access

    A Socio-Epistemic Theory of Attachment: Part III. A Relevance-Based Taxonomy for the Adult Attachment Interview (AAI) Coding System

    by Alessandro Talia, Robbie Duschinsky, Diana Mazzarella, Katarina Begus, Johann R. Kleinbub, Madeleine Miller-Bottome, Svenja Taubner, Jeremy Holmes, Peter Fonagy

    In the third and last part of this article, we consider a possible extension of the socio-epistemic theory of attachment- related differences, and we apply the theory to other previously underexplored aspects of the Adult Attachment Interview (AAI) coding system. Building on our earlier proposal that attachment patterns reflect differences in epistemic trust, we examine how this framework can illuminate three further key areas of AAI classification: (1) the AAI subclassifications (2) the unresolved/disorganized (U/d) classification; and (3) additional features typical of speakers who are assigned the secure/autonomous AAI classification, including mentalizing. By integrating Relevance Theory, we argue that these dimensions of variation reflect epistemic styles, or systematic differences in how individuals evaluate and share knowledge within interpersonal exchanges. This approach refines our understanding of the AAI coding system and further supports a possible theoretical shift towards conceptualizing attachment differences as fundamental variations in socio-epistemic processes. Keywords: attachment theory, Adult Attachment Interview, mentalizing, trauma, pragmatics, Relevance theory