Trends in Cognitive Sciences
ReviewWhither geometry? Troubles of the geometric module
Section snippets
Modularity of geometry in doubt
In the 1980s, modularity of mind was an influential proposal 1, 2. Fodor [2] proposed that peripheral perceptual processes are largely modular, but that central cognitive processes are largely nonmodular. Out of the flourishing study of spatial cognition in rats at the time 3, 4, 5, 6, a case of central modularity was proposed with the discovery of rotational errors in rats 7, 8. In these errors, rats confused diagonally opposite locations in a rectangular arena (Box 1). The phenomenon
Cue competition
Cue competition is a key concept in animal learning [28]. When multiple cues can be learned, one cue might interfere with the learning of another. When both Cue A and Cue B presented together predict an outcome, A might be less well learned than if it had been presented alone as the predictor. This phenomenon is called overshadowing [29]. Alternatively, a prior history of learning Cue B as a predictor might result in slower learning of Cue A when Cue A and Cue B are subsequently presented
If only they learned geometric cues: obligatory learning in doubt
In an important finding, rats had trouble learning to use geometric cues in a task in which it would have helped greatly to do so. Let us call this the color flip-flop task because the colors of walls changed frequently in the kite-shaped pool during training [22]. One side of the kite had a long black wall adjacent to a short black wall, the other side had a long white wall adjacent to a short white wall. But which color was on which side (top or bottom walls in Figure 1b) changed at random
Associative model of geometry and feature learning
How can features sometimes help the learning of geometry and sometimes hinder it? A recent associative model adapted the famous Rescorla-Wagner model [33] for geometry studies [23] (Box 2). The Rescorla-Wagner model was designed to account for cue competition in experiments on classical conditioning, in which all important events are determined by the experimenter. Events include one unconditioned stimulus of biological significance such as the presentation of food or electric shock and
View-based matching
A different attempt to deal with the troubles of the geometric module is view-based matching, a scheme that requires neither explicit extraction of geometric properties nor, a fortiori, a geometric module. The idea is not new at all; it has had a long tradition in insect navigation 34, 35, 36, 37, 38. The basic idea is to take a ‘snapshot’ of the scene surrounding the goal. The snapshot becomes the target view to recover, and the animal moves so as to reduce the discrepancy between the current
Whither geometry? Some speculations
Research on rats has a large influence on the study of human spatial cognition. In an exciting recent reformulation of geometry and features, phenomena of cue competition play a crucial part in delineating systems of spatial representation 26, 27. In virtual search tasks on a computer monitor, cue competition was found when humans located objects on the basis of nearby individual landmarks. By contrast, cue competition was absent when humans located objects with reference to the boundaries of
Concluding remarks
Several claims about geometry and features in Cheng and Newcombe's review [14] have been overturned by recent evidence in rats. Recent work shows that geometry and features are not always learned independently. Interactions in the form of cue competition and potentiation are rife, with large features hindering or sometimes helping the learning of geometric cues. Specifying the conditions favoring cue competition, independent learning or potentiation is an important empirical problem on which
Acknowledgements
I thank Anthony McGregor, Sara Shettleworth, Brett Gibson, Jochen Zeil and two anonymous reviewers for helpful comments on an earlier draft of this article.
References (70)
Spatial localization does not require the presence of local cues
Learn. Motiv.
(1981)A purely geometric module in the rat's spatial representation
Cognition
(1986)- et al.
Modularity and development: the case of spatial reorientation
Cognition
(1996) Toddler's use of metric information and landmarks to reorient
J. Exp. Child Psychol.
(2001)Sources of flexibility in human cognition: dual-task studies of space and language
Cognit. Psychol.
(1999)Pigeon homing along highways and exits
Curr. Biol.
(2004)Geometric determinants of spatial memory
Cognition
(2004)Small-scale spatial cognition in pigeons
Behav. Processes
(2006)- et al.
Human spatial representation: insights from animals
Trends Cogn. Sci.
(2002) - et al.
Is language necessary for human spatial reorientation? reconsidering evidence from dual task paradigms
Cognit. Psychol.
(2008)
Spatial memory: how egocentric and allocentric combine
Trends Cogn. Sci.
Reorientation in a rhombic environment: no evidence for an encapsulated geometric module
Cogn. Dev.
Rules and Representations
Modularity of Mind
Remembrance of places passed: spatial memory in rats
J. Exp. Psychol. Anim. Behav. Process.
The Hippocampus as a Cognitive Map
The Organization of Learning
Heading in the rat: determination by environmental shape
Anim. Learn. Behav.
Modularity as a fish (Xenotoca eiseni) views it: conjoining geometric and nongeometric information for spatial reorientation
J. Exp. Psychol. Anim. Behav. Process.
Encoding of geometric and featural spatial information by goldfish (Carassius auratus)
J. Comp. Psychol.
Geometric modules in animals’ spatial representations: a test with chicks (Gallus gallus domesticus)
J. Comp. Psychol.
Pigeons’ (Columba livia) encoding of geometric and featural properties of a spatial environment
J. Comp. Psychol.
Rhesus monkeys use geometric and nongeometric information during a reorientation task
J. Exp. Psychol. Gen.
Is there a geometric module for spatial orientation? Squaring theory and evidence
Psychon. Bull. Rev.
A geometric process for spatial reorientation in young children
Nature
Children's use of landmarks: implications for modularity theory
Psychol. Sci.
Spatial encoding in mountain chickadees: features overshadow geometry
Biol. Lett.
Potentiation, overshadowing, and blocking of spatial learning based on the shape of the environment
J. Exp. Psychol. Anim. Behav. Process.
Spatial learning based on the shape of the environment is influenced by properties of the objects forming the shape
J. Exp. Psychol. Anim. Behav. Process.
An associative model of geometry learning
J. Exp. Psychol. Anim. Behav. Process.
Information content of panoramic images: I. rotational errors and the similarity of views in rectangular arenas
J. Exp. Psychol. Anim. Behav. Process.
Information content of panoramic images: II. view-based navigation in nonrectangular experimental arenas
J. Exp. Psychol. Anim. Behav. Process.
Distinct error-correcting and incidental learning of location relative to landmarks and boundaries
Proc. Natl. Acad. Sci. U. S. A.
Parallel striatal and hippocampal systems for landmarks and boundaries in spatial memory
Proc. Natl. Acad. Sci. U. S. A.
Theories of associative learning in animals
Annu. Rev. Psychol.
Cited by (175)
A map of spatial navigation for neuroscience
2023, Neuroscience and Biobehavioral ReviewsYoung children's representation of geometric relationships between locations in location coding
2020, Journal of Experimental Child PsychologyUse of medial axis for reorientation by the Clark's nutcracker (Nucifraga columbiana)
2019, Behavioural Processes