skip to main content

Title: Strategies for visuospatial reasoning: Experiments in sufficiency and diversity
In this paper, we present the Visuospatial Reasoning Environment for Experimentation (VREE). VREE provides a simulated environment where intelligent agents interact with virtual objects while solving different visuospatial reasoning tasks. This paper shows how VREE is valuable for studying the sufficiency of visual imagery approaches for a large number of visuospatial reasoning tasks as well as how diverse strategies can be represented and studied within a single task. We present results from computational experiments using VREE on the block design task and on numerous subtests from the Leiter-R test battery on nonverbal intelligence.  more » « less
Award ID(s):
1730044 1922697
Author(s) / Creator(s):
; ; ; ;
Date Published:
Journal Name:
Proceedings of the Eighth Annual Conference on Advances in Cognitive Systems (ACS)
Medium: X
Sponsoring Org:
National Science Foundation
More Like this
  1. This paper presents a systematic review of the empirical literature that uses dual-task interference methods for investigating the on-line involvement of language in various cognitive tasks. In these studies, participants perform some primary task X putatively recruiting linguistic resources while also engaging in a secondary, concurrent task. If performance on the primary task decreases under interference, there is evidence for language involvement in the primary task. We assessed studies (N = 101) reporting at least one experiment with verbal interference and at least one control task (either primary or secondary). We excluded papers with an explicitly clinical, neurological, or developmental focus. The primary tasks identified include categorization, memory, mental arithmetic, motor control, reasoning (verbal and visuospatial), task switching, theory of mind, visual change, and visuospatial integration and wayfinding. Overall, the present review found that internal language is likely to play a facilitative role in memory and categorization when items to be remembered or categorized have readily available labels, when inner speech can act as a form of behavioral self-cuing (inhibitory control, task set reminders, verbal strategy), and when inner speech is plausibly useful as “workspace,” for example, for mental arithmetic. There is less evidence for the role of internal language in cross-modal integration, reasoning relying on a high degree of visual detail or items low on nameability, and theory of mind. We discuss potential pitfalls and suggestions for streamlining and improving the methodology. 
    more » « less
  2. Observations abound about the power of visual imagery in human intelligence, from how Nobel prize-winning physicists make their discoveries to how children understand bedtime stories. These observations raise an important question for cognitive science, which is, what are the computations taking place in someone’s mind when they use visual imagery? Answering this question is not easy and will require much continued research across the multiple disciplines of cognitive science. Here, we focus on a related and more circumscribed question from the perspective of artificial intelligence (AI): If you have an intelligent agent that uses visual imagery-based knowledge representations and reasoning operations, then what kinds of problem solving might be possible, and how would such problem solving work? We highlight recent progress in AI toward answering these questions in the domain of visuospatial reasoning, looking at a case study of how imagery-based artificial agents can solve visuospatial intelligence tests. In particular, we first examine several variations of imagery-based knowledge representations and problem-solving strategies that are sufficient for solving problems from the Raven’s Progressive Matrices intelligence test. We then look at how artificial agents, instead of being designed manually by AI researchers, might learn portions of their own knowledge and reasoning procedures from experience, including learning visuospatial domain knowledge, learning and generalizing problem-solving strategies, and learning the actual definition of the task in the first place.

    more » « less
  3. null (Ed.)
    Visuospatial reasoning refers to a diverse set of skills that involve thinking about space and time. An artificial agent with access to a sufficiently large set of visuospatial reasoning skills might be able to generalize its reasoning ability to an unprecedented expanse of tasks including portions of many popular intelligence tests. In this paper, we stress the importance of a developmental approach to the study of visuospatial reasoning, with an emphasis on fundamental skills. A comprehensive benchmark, with properties we outline in this paper including breadth, depth, explainability, and domain-specificity, would encourage and measure the genesis of such a skillset. Lacking an existing benchmark that satisfies these properties, we outline the design of a novel test in this paper. Such a benchmark would allow for expanding analysis of existing datasets’ and agents’ applicability to the problem of generalized visuospatial reasoning. 
    more » « less
  4. Large language models (LLMs) have demonstrated an impressive ability to perform arithmetic and symbolic reasoning tasks, when provided with a few examples at test time ("few-shot prompting"). Much of this success can be attributed to prompting methods such as "chain-of-thought", which employ LLMs for both understanding the problem description by decomposing it into steps, as well as solving each step of the problem. While LLMs seem to be adept at this sort of step-by-step decomposition, LLMs often make logical and arithmetic mistakes in the solution part, even when the problem is decomposed correctly. In this paper, we present Program-Aided Language models (PAL): a novel approach that uses the LLM to read natural language problems and generate programs as the intermediate reasoning steps, but offloads the solution step to a runtime such as a Python interpreter. With PAL, decomposing the natural language problem into runnable steps remains the only learning task for the LLM, while solving is delegated to the interpreter. We demonstrate this synergy between a neural LLM and a symbolic interpreter across 13 mathematical, symbolic, and algorithmic reasoning tasks from BIG-Bench Hard and others. In all these natural language reasoning tasks, generating code using an LLM and reasoning using a Python interpreter leads to more accurate results than much larger models. For example, PAL using Codex achieves state-of-the-art few-shot accuracy on GSM8K, surpassing PaLM which uses chain-of-thought by absolute 15% top-1. 
    more » « less
  5. Recent years have seen the rapid adoption of artificial intelligence (AI) in every facet of society. The ubiquity of AI has led to an increasing demand to integrate AI learning experiences into K-12 education. Early learning experiences incorporating AI concepts and practices are critical for students to better understand, evaluate, and utilize AI technologies. AI planning is an important class of AI technologies in which an AI-driven agent utilizes the structure of a problem to construct plans of actions to perform a task. Although a growing number of efforts have explored promoting AI education for K-12 learners, limited work has investigated effective and engaging approaches for delivering AI learning experiences to elementary students. In this paper, we propose a visual interface to enable upper elementary students (grades 3-5, ages 8-11) to formulate AI planning tasks within a game-based learning environment. We present our approach to designing the visual interface as well as how the AI planning tasks are embedded within narrative-centered gameplay structured around a Use-Modify-Create scaffolding progression. Further, we present results from a qualitative study of upper elementary students using the visual interface. We discuss how the Use-Modify-Create approach supported student learning as well as discuss the misconceptions and usability issues students encountered while using the visual interface to formulate AI planning tasks. 
    more » « less