# Toward a universal law of generalization for psychological science.

@article{Shepard1987TowardAU, title={Toward a universal law of generalization for psychological science.}, author={Roger N. Shepard}, journal={Science}, year={1987}, volume={237 4820}, pages={ 1317-23 } }

A psychological space is established for any set of stimuli by determining metric distances between the stimuli such that the probability that a response learned to any stimulus will generalize to any other is an invariant monotonic function of the distance between them. To a good approximation, this probability of generalization (i) decays exponentially with this distance, and (ii) does so in accordance with one of two metrics, depending on the relation between the dimensions along which the… Expand

#### Topics from this paper

#### 2,176 Citations

The Generalized Universal Law of Generalization

- Computer Science, Mathematics
- ArXiv
- 2001

It is shown that the Universal Law of Generalization holds with probability going to one—provided the probabilities concerned are computable, and a mathematically more appealing form of theUniversal Law is given. Expand

Measurement invariance explains the universal law of generalization for psychological perception

- Mathematics, Biology
- 2018

It is shown that the few inevitable invariances that must apply to any reasonable perceptual scale provide a sufficient explanation for the universal exponential law of generalization. Expand

Efficient coding explains the universal law of generalization in human perception

- Computer Science, Medicine
- Science
- 2018

It is shown that the universal law emerges inevitably from any information processing system (whether biological or artificial) that minimizes the cost of perceptual error subject to constraints on the ability to process or transmit information. Expand

Measurement invariance explains the universal law of generalization for psychological perception

- Mathematics, Medicine
- Proceedings of the National Academy of Sciences
- 2018

The few inevitable invariances that must apply to any reasonable perceptual scale provide a sufficient explanation for the universal exponential law of generalization, which is shown to be invariant to shift by a constant value which by itself leads to the exponential form. Expand

Connectionist Implementation of a Theory of Generalization

- Mathematics, Computer Science
- NIPS
- 1990

A natural connectionist implementation of the generalization theory is described, and how implications of the theory for generalization, discrimination, and classification learning can be explored by connectionist simulation is illustrated. Expand

On the dynamics of generalization.

- Medicine
- Psychological review
- 1990

Shepard (1987) has proposed a universal law for stimulus generalization that deals with relations in a "psychological space," established through multidimensional curve-fitting techniques (MDS), and process theories to reconcile the Gaussian and exponential data. Expand

On the dynamics of generalization.

- Mathematics
- 1990

Stimulus generalization is the tendency, after experience with a given stimulus, to react in the same way to other, similar stimuli. Shepard (1987) has proposed a universal law forstimulus… Expand

Generalization, similarity, and Bayesian inference.

- Medicine, Psychology
- The Behavioral and brain sciences
- 2001

Here Shepard's theory is recast in a more general Bayesian framework and it is shown how this naturally extends his approach to the more realistic situation of generalizing from multiple consequential stimuli with arbitrary representational structure. Expand

Shepard's Universal Law Supported by Honeybees in Spatial Generalization

- Psychology, Medicine
- Psychological science
- 2000

Support in an invertebrate animal increases the scope of Shepard's law, and suggests that the ecological structure of the world may have driven the evolution of cognitive structures in diverse animals. Expand

From natural kinds to complex categories

- Computer Science
- 2006

The rational analyses of generalization proposed by Shepard (1987) and Tenenbaum and Griffiths (2001a) are extended to category learning and selective attention and typicality gradients are shown to emerge from Bayesian inference over the possible category microstructures and task contexts. Expand

#### References

SHOWING 1-10 OF 62 REFERENCES

Stimulus and response generalization: A stochastic model relating generalization to distance in psychological space

- Mathematics
- 1957

A mathematical model is developed in an attempt to relate errors in multiple stimulus-response situations to psychological inter-stimulus and inter response distances. The fundamental assumptions are… Expand

Stimulus and response generalization: tests of a model relating generalization to distance in psychological space.

- Mathematics, Medicine
- Journal of experimental psychology
- 1958

A mathematical model is developed in an attempt to relate errors in multiple stimulus-response situations to psychological inter-stimulus and inter response distances. The fundamental assumptions are… Expand

Attention and the metric structure of the stimulus space.

- Mathematics
- 1964

Abstract Three experiments were performed in an investigation of how differences in size and inclination combine to determine the over-all similarity between otherwise identical visual stimuli.… Expand

Perceptual-cognitive explorations of a toroidal set of free-form stimuli

- Mathematics
- 1973

Abstract A polar-coordinate analogue of Fourier synthesis generates organic-appearing “free forms” that can be continuously deformed along any desired number of difficult-to-verbalize dimensions.… Expand

Metric structures in ordinal data

- Mathematics
- 1966

Abstract Under appropriate conditions, data merely about the ordering of objects—or of the separations between objects—is sometimes sufficient to fix the positions of those objects on an essentially… Expand

Monotone mapping of similarities into a general metric space

- Mathematics
- 1974

Abstract A method of “maximum variance nondimensional scaling” is described and tested that transforms similarity measures into distances that meet just three conditions: (C1) they exactly satisfy… Expand

The analysis of proximities: Multidimensional scaling with an unknown distance function. II

- Mathematics
- 1962

The first in the present series of two papers described a computer program for multidimensional scaling on the basis of essentially nonmetric data. This second paper reports the results of two kinds… Expand

Representation of the orientations of shapes.

- Computer Science, Medicine
- Acta psychologica
- 1985

It is suggested that the constraints governing possible transformations of objects have become so internalized that the authors unconsciously represent these possible motions upon the visual presentation of any object. Expand

Multidimensional scaling by optimizing goodness of fit to a nonmetric hypothesis

- Mathematics
- 1964

Multidimensional scaling is the problem of representingn objects geometrically byn points, so that the interpoint distances correspond in some sense to experimental dissimilarities between objects.… Expand