United States

Imitation is a key component of human social behavior, and is widely used by both children and adults as a way to navigate uncertain or unfamiliar situations. But in an environment populated by multiple heterogeneous agents pursuing different goals or objectives, indiscriminate imitation is unlikely to be an effective strategy---the imitator must instead determine who is most useful to copy. There are likely many factors that play into these judgements, depending on context and availability of information. Here we investigate the hypothesis that these decisions involve inferences about other agents&#39; reward functions. We suggest that people preferentially imitate the behavior of others they deem to have similar reward functions to their own. We further argue that these inferences can be made on the basis of very spare or indirect data, by leveraging an inductive bias toward positing the existence of different groups or types of people with similar reward functions, allowing learners to select imitation targets without direct evidence of alignment.

CogSci 2023

Selective imitation on the basis of reward function similarity

The purpose of the AAAI conference series is to promote research in Artificial Intelligence (AI) and foster scientific exchange between researchers, practitioners, scientists, students, and engineers across the entirety of AI and its affiliated disciplines. AAAI-23 is the Thirty-Seventh AAAI Conference on Artificial Intelligence. The theme of this conference is to create collaborative bridges within and beyond AI. Like previous AAAI conferences, AAAI-23 will feature technical paper presentations, special tracks, invited speakers, workshops, tutorials, poster sessions, senior member presentations, competitions, and exhibit programs, and two new activities: a Bridge Program and a Lab Program. Many of these activities are tailored to the theme of bridges and all are selected according to the highest standards, with additional programs for students and young researchers. 
AAAI is providing you with a conference planner, which you can use to help organize your itinerary of activities. This includes talks to attend in person, talks to attend remotely, breaks with colleagues and your site seeing activities. To access this conference planner, please go to [https://aaai-2023.takemobi.io](https://aaai-2023.takemobi.io).

In order to access this site, you need to register. If you haven't already, please register [here](https://aaai.org/Conferences/AAAI-23/registration/).


AAAI 2023

The purpose of the AAAI conference series is to promote research in Artificial Intelligence (AI) and foster scientific exchange between researchers, practitioners, scientists, students, and engineers across the entirety of AI and its affiliated disciplines.

Imitation is a key component of human social behavior, and is widely used by both children and adults as a way to navigate uncertain or unfamiliar situations. But in an environment populated by multiple heterogeneous agents pursuing different goals or objectives, indiscriminate imitation is unlikely to be an effective strategy---the imitator must instead determine who is most useful to copy. There are likely many factors that play into these judgements, depending on context and availability of information. Here we investigate the hypothesis that these decisions involve inferences about other agents' reward functions. We suggest that people preferentially imitate the behavior of others they deem to have similar reward functions to their own. We further argue that these inferences can be made on the basis of very spare or indirect data, by leveraging an inductive bias toward positing the existence of different groups or types of people with similar reward functions, allowing learners to select imitation targets without direct evidence of alignment.

poster

The 45th Annual Meeting of the Cognitive Science Society will take place in Sydney, Australia as a modified hybrid conference. Presentations from the main-session room will be live-streamed every day for all events. This will include all invited talks and award sessions. Recordings of all CogSci 2023 sessions will be made available for on-demand viewing after the conference. 
The in-person component will be held at the International Convention Centre Sydney at Darling Harbour. The Conference will take place Wednesday July 26th – Saturday July 29th, 2023 (Wednesday, July 26th is the Workshops Day).

Join us!

 [**The Program at a Glance can be found here**](https://cognitivesciencesociety.org/wp-content/uploads/2023/07/COgSci-PAAG-V2.pdf)
 [**The Detailed Program can be found here**](https://online.fliphtml5.com/ebtyf/mswz/)
 

#### **CogSci 2023 Program Book**
<iframe style="width:700px;height:400px" src="https://online.fliphtml5.com/ebtyf/tzgk/" seamless="seamless" scrolling="no" frameborder="0" allowtransparency="true" allowfullscreen="true" ></iframe>

The 45th Annual Meeting of the Cognitive Science Society presents the latest research across cognitive science and highlights the theme of Cognition in Context.

The human brain can quickly organise visually perceived objects into categories. Although previous research has focussed on object and category representations in the brain, it is still unclear how one category of ecological importance, food, is represented in the brain. Here, we used time-resolved multivariate analyses of electroencephalography data to investigate the time-course of food representations in the brain. Our results show that the brain distinguishes between food and non-food items from approximately 84 ms onwards. The neural signal also contained information about the naturalness, the level of transformation, as well as the perceived caloric content of food. In addition, the recorded brain activity predicted the behavioural responses of different groups of participants who participated in one of two tasks: a naturalness food categorisation task and an odd-item-out task on 3 presented food items. Together, our results contribute to our understanding of how the human brain processes visually presented food.

The representational dynamics of food in the human brain

From learning to play video games to using novel tools, humans are able to acquire a variety of complex mappings between their actions and arbitrary outcomes. In addition, once they have learned such mappings, they often have to use them sequentially to achieve goals, i.e., planning. In this work we study how the learning of a novel mapping interacts with planning in the context of grid navigation. In order to do so, we developed a computer-based game where subjects have to move a cursor from start to target locations using the keys of their keyboard. Importantly, to more closely resemble the complexity of the mappings that people acquire in their lives, the cursor movement was determined by a non-trivial rule inspired by the movement of the piece of chess known as the Knight. In Experiment 1, we show that participants were able to improve their performance in our task, though not arriving optimally to the targets in the majority of trials. Additionally, we proposed two computational models where the mapping-learning and planning interaction occurs in the form of an action value computation, and then evaluated their fit to our data. Finally, in Experiment 2, we showed that exposing participants to the mapping component of the task without having to plan, provides a performance improvement when exposed to the full task later. Crucially, this improvement does not occur if subjects are exposed to the planning component of the task prior to doing it fully.


Exploring human learning and planning on grid navigation using arbitrary mappings.

Physical assembly is a difficult planning problem. Humans do it efficiently by breaking large problems into smaller, easier to solve portions. But what governs which portions are chosen? 
We present a computational model that predicts that humans break physical assembly problems down to minimize cognitive costs.
We test this by asking participants to choose which part of a tower they want to build next. 
Participants reliably choose the easier to solve subgoal out of two otherwise similar options. 
Beyond the immediate cognitive cost, participants also consider how difficult the rest of the tower will be to solve.
A model that takes into account near-future cognitive costs best predicts participants' choices.
These findings show that humans can estimate how difficult solving a subgoal will be,
and that they choose subgoals to minimize immediate and future cognitive costs.
These results help explain how humans make efficient use of cognitive resources to solve complex planning problems.

Humans choose visual subgoals to reduce cognitive cost

This study explores eye-markers for sub-categories of cognitive load. Experiments were conducted on 63 participants using Image Sliding Puzzle (ISP). NASA-TLX was administered post task completion as a measure of cognitive load. Total scanning duration, total fixation duration, fixation count and total saccadic duration were found to be significant, which is consistent with pre-existing literature. Next, we investigated whether sub-categories of cognitive load (mental demand, temporal demand, perceived performance, effort, and frustration) can be distinguished by characteristic eye-metrics. Our findings reveal signature eye-markers for specific sub-categories of cognitive load. Further, we explored the link between perceived performance and actual performance and established that mean fixation duration, peak velocity, mean saccadic duration, and skewness in saccadic velocity were significant markers for both objective and subjective markers of performance. To our knowledge this is the first study to compare the task-evoked eye measures for sub-categories of cognitive load.


Estimating Sub-categories of Cognitive Load: An Eye-tracking Study

Active Inference and its accompanying Bayesian Mechanics
(BM) are important psychological and cognitive science theories.
While there is a strong interaction between the theories
and the philosophical realm, it needs to be clarified what its
metaphysical commitments are. We tease out these commitments
while looking from the perspective of the psychology
of goals. We find that Active Inference cannot account for
the dynamic growth of goals, primarily because of its closed
generative model. We trace the reason for this through the
fixed ‘ontological constraint’ of BM to the inherent ‘logic of
machines’ in mechanistic models, grounding finally in the incompatibility
between substance metaphysics, impredicativity,
and emergence. Thus we argue that Active Inference implicitly
presupposes a Substance metaphysics, yielding the theory no
resource to model novelty, growth, and development observed
in human psychology. We briefly sketch a powerful alternative
grounded in process metaphysics to model biological and
cognitive systems.


Active Inference and Psychology of Goals: A study in Substance and Process Metaphysics

People are often confronted with problems whose complexity exceeds their cognitive capacities. To deal with this complexity, individuals and managers can break complex problems down into a series of subgoals. Which subgoals are most effective depends on people's cognitive constraints and the cognitive mechanisms of goal pursuit. This creates an untapped opportunity to derive practical recommendations for which subgoals managers and individuals should set from cognitive models of bounded rationality. To seize this opportunity, we apply the principle of resource rationality to formulate a mathematically precise normative theory of (self-)management by goal-setting. We leverage this theory to computationally derive optimal subgoals from a resource-rational model of human goal pursuit. Finally, we show that the resulting subgoals improve the problem-solving performance of bounded agents and human participants. This constitutes a first step towards grounding prescriptive theories of management and practical recommendations for goal-setting in computational models of the relevant psychological processes and cognitive limitations.


Toward a normative theory of (self-)management by goal-setting

Cumulative culture requires efficient learning mechanisms that can withstand environmental change across generations. We unify two competing theories of the learning mechanism supporting cumulative culture in a common computational framework, distinguishing model-based from model-free social learning. We describe and analyze evolutionary models that explain when and why model-based and model-free social learning are each optimal, and in particular how environmental determines which strategy succeeds. Strikingly, we find that model-based social learning can succeed even in high-volatility environments. These results yield novel predictions concerning cultural variation in social learning mechanisms.


Computational principles underlying the evolution of cultural learning mechanisms

The expressive power of natural languages depends on their regular compositional structure, which allows us to express and understand an infinite set of messages. However, a complete model of language evolution should also account for irregular exceptions to regular rules, common in natural languages. Historical linguistics has established a correlation between irregularity and frequency in language use, which has been attributed to preferential irregularisation of frequent items, or preferential regularisation of infrequent items. In an iterated learning experiment where participants learn and reproduce a miniature language across multiple generations, we show that this correlation can be explained by the relationship between frequency, regularity and learnability, without needing to appeal to frequency-dependent irregularisation. We find that systems of plural marking regularise across generations of transmission, but that high-frequency items remain irregular. Our results further show that the persistence of irregularity is due to high frequency overriding pressures which normally reduce learnability.


The Relationship Between Frequency and Irregularity in the Evolution of Linguistic Structure: An Experimental Study

Many people turn to social media for public health information, 
but such platforms often contain conflicting and inaccurate 
medical advice. To assess complex health claims online, 
people may consider the prevailing consensus; however, 
previous work suggests that people may not be very sensitive 
to important cues to consensus “quality”. To explore further, 
across two experiments we tested people’s sensitivity to the 
consensus-quality cues of source diversity and source 
expertise. Via a mock Twitter platform, participants rated their 
belief in a series of health claims both before and after reading 
various kinds of tweets about the claims. Experiment 1 showed 
that experts (both individual medical experts and health 
organisations) were more persuasive than non-experts. 
Additionally, stances that were supported by a diverse set of 
sources were more persuasive. Experiment 2 showed that 
participants continue to favour experts even when 
outnumbered in tweet quantity by non-experts. When experts 
were not present, however, participants favoured high tweet 
quantity. Both experiments suggest that cues to consensus 
quality (namely, expertise and source diversity) and consensus 
quantity (tweet quantity) are salient cues in belief revision. 
These findings are important in understanding how socially 
acquired health information (and misinformation) shifts 
opinion, and the role that experts can play.


The Influence of Cues to Consensus Quantity and Quality on Belief in Health Claims

Consensus-based social learning strategies often outcompete other strategies in evolutionary models. But while formal proofs suggest that consensus’ reliability is compromised when individual judgments are not independent, this makes for a notoriously implausible assumption in the biological world: the people we learn from are constantly learning from each other as well. How do we avoid being misled by consensus? We present two experiments and a computational model examining commonsense reasoning about how people’s public and private judgments are influenced by the consensus and social status of those around them. Results suggest that while people realize that these two factors can cause others’ public and private judgments to diverge, their own trust in public consensus depends on how accurately they believe it reflects their informants’ true beliefs.


Agenda setting and The Emperor’s New Clothes: people infer that letting powerful agents make their opinion known early can trigger information cascades and pluralistic ignorance

Premium content

Downloads

Next from CogSci 2023

The representational dynamics of food in the human brain

Similar lecture

Co-Imitation: Learning Design and Behaviour by Imitation