Gestalt Reasoning with Conjunctions and Disjunctions

RESEARCH ARTICLE Gestalt Reasoning with Conjunctions and Disjunctions Magda L. Dumitru1,2*, Gitte H. Joergensen3,4 1 Department of Cognitive Science, Macquarie University, Sydney, NSW, Australia, 2 Department of Cognitive Science, METU, Ankara, Turkey, 3 Department of Psychology, University of York, Heslington, York, United Kingdom, 4 Department of Psychology, University of Connecticut, Storrs, CT, United States of America * Abstract OPEN ACCESS Citation: Dumitru ML, Joergensen GH (2016) Gestalt Reasoning with Conjunctions and Disjunctions. PLoS ONE 11(3): e0151774. doi:10.1371/journal. pone.0151774 Editor: Ulrich von Hecker, Cardiff University, UNITED KINGDOM Received: September 4, 2015 Accepted: March 3, 2016 Published: March 17, 2016 Copyright: © 2016 Dumitru, Joergensen. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. Data Availability Statement: Data from the three studies reported here are available at https://osf.io/ 4g62y/ DOI 10.17605/OSF.IO/AYNCZ | ARK c7605/ osf.io/ayncz. Funding: Magda L. Dumitru was supported by a Marie Curie fellowship FP7-PEOPLE-2013-IAPP (grant number 610986). The funder had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. Competing Interests: The authors have declared that no competing interests exist. Reasoning, solving mathematical equations, or planning written and spoken sentences all must factor in stimuli perceptual properties. Indeed, thinking processes are inspired by and subsequently fitted to concrete objects and situations. It is therefore reasonable to expect that the mental representations evoked when people solve these seemingly abstract tasks should interact with the properties of the manipulated stimuli. Here, we investigated the mental representations evoked by conjunction and disjunction expressions in language-picture matching tasks. We hypothesised that, if these representations have been derived using key Gestalt principles, reasoners should use perceptual compatibility to gauge the goodness of fit between conjunction/disjunction descriptions (e.g., the purple and/ or the green) and corresponding binary visual displays. Indeed, the results of three experimental studies demonstrate that reasoners associate conjunction descriptions with perceptuallydependent stimuli and disjunction descriptions with perceptually-independent stimuli, where visual dependency status follows the key Gestalt principles of common fate, proximity, and similarity. Introduction People have many options available for solving complex tasks in real-life situations. One option is to apply the rules of propositional logic [1–4] and identify the relations (e.g., cause, consequence, condition, coordination) holding between the main action to be performed and further associated actions. For example, a medical emergency protocol recommends US service providers that “If no shock is indicated, consider termination of resuscitation protocol or transport patient immediately”. To evaluate this statement, people need to combine the truth-values of the coordination statement (composed of two clauses linked by the disjunction connective ‘or’) and of the conditional statement (introduced by the connective ‘if’). Specifically, disjunction statements are true unless both parts fail to obtain (i.e., resuscitation continues and patient is not transported) and conditional statements are true unless the first part obtains while the second part does not (i.e., there is no shock indicated, yet resuscitation continues and patient is PLOS ONE | DOI:10.1371/journal.pone.0151774 March 17, 2016 1 / 17 Gestalt Reasoning not transported). Nevertheless, experimental evidence increasingly suggests that naive reasoners forego formal rules and instead draw on general cognitive mechanisms when processing complex information. Accordingly, their responses are often modulated by stimuli perceptual properties [5], by metaphors that ground incoming information onto basic action patterns [6– 7], or by mental models that represent, in analogical form, key elements in a message and the relationships between them [8–10]. The present paper builds on previous findings [5] suggesting that a general mechanism of information processing, namely parsing (aka ‘chunking’) shapes reasoning with conjunctions and disjunctions. While there is ample experimental evidence that incoming information is often chunked into smaller units to facilitate processing [11–12] the nature and properties of these chunks are largely unknown. Here, we advance the hypothesis that, when processing conjunctions and disjunctions, reasoners instantly build Gestalt-like representations (one and two mental objects respectively—cf. [5]), which generate perceptual compatibility effects predicted by key Gestalt principles (i.e., common fate, proximity, and similarity). As such, Gestalt representations of conjunctions and disjunctions are not unlike concrete object representations stored in long-term memory and therefore they should interact with the properties of the perceptual stimuli that reasoners manipulate when solving a particular task. The remainder of this paper is organised as follows. After summarising the main tenets of Gestalt theory, we outline the advantages of the earliest analogy-based approaches, in particular the important progress made by model theory towards achieving a psychologically plausible account of conjunction and disjunction representations. Subsequently, we present three experimental studies investigating a novel account according to which reasoners build Gestalt-like conjunction and disjunction representations in online language-picture matching tasks. We expect that, unlike previous reasoning accounts, Gestalt reasoning based on indirect analogies can successfully predict the observed response patterns. Gestalt psychologists ever since Koehler [13], Koffka [14], and Wertheimer [15] have developed a phenomenological appraisal of perceptual processes according to which “the whole is different from the sum of its parts”. In their view, perceptual information must go well beyond stimuli properties or bodily sensations for people to be able to group them into meaningful units, which facilitate mental processes and attention allocation. Moreover, evidence from several cognitive domains (for reviews see Tversky [16–17]) demonstrates that grouping is a general process of the mind, hence is not limited to perception. Studies of cognitive maps, for instance, have revealed that grouping can readily occur in memory, as individuals were faster to determine which of a pair of cities was farther east or farther north when the cities were located in different states or countries than when they were in the same state or country [18–19]. Likewise, people tend (...truncated)