A feature-free method for testing adaptive network learning models is presented. The test is based on a property called the mean matching law, a. property shared by many adaptive network models of learning. As an application of this method, we prove that cue competition effects obtained with statistically independent cues cannot be explained by many previous adaptive network learning models, including those based on the delta learning rule. These results point to the need to incorporate competitive learning properties into adaptive network learning models.
Get full access to this article
View all access options for this article.
References
1.
AndersonJ.A.SilversteinJ.W.RitzS.A.JonesR.S. (1977). Distinctive features, categorical perception, and probability learning: Some applications of a neural model. Psychological Review, 84, 413–451.
2.
BirnbaumM.H. (1976). Intuitive numerical prediction. American Journal of Psychology, 89, 417–429.
3.
BusemeyerJ.R.MyungI.J. (1988). A new method for investigating prototype learning. Journal of Experimental Psychology: Learning, Memory, and Cognition, 14, 3–11.
4.
ChapmanG.B. (1991). Trial order affects cue interaction in contingency judgment. Journal of Experimental Psychology: Learning, Memory, and Cognition, 17, 837–854.
5.
EstesW.K.CampbellJ.A.HatsopoulosN.HurwitzJ.B. (1989). Base-rate effects in category learning: A comparison of parallel network and memory storage-retrieval models. Journal of Experimental Psychology: Learning, Memory, and Cognition, 15, 556–571.
6.
EstesW.K.StraughanJ.H. (1954). Analysis of a verbal conditioning situation in terms of statistical learning theory. Journal of Experimental Psychology, 47, 225–234.
7.
GluckM.A. (1991). Stimulus generalization and representation in adaptive network models of category learning. Psychological Science, 2, 50–55.
8.
GluckM.A.BowerG.H. (1988a). Evaluating an adaptive network model of human learning. Journal of Memory and Language, 27, 166–195.
9.
GluckM.A.BowerG.H. (1988b). From conditioning to category learning: An adaptive network model. Journal of Experimental Psychology: General, 117, 227–247.
10.
GrossbergS. (1982). Studies of mind and brain: Neural principles of learning, perception, development, cognition, and motor control. Boston: Reidel Press.
11.
GrossbergS. (1987). Competitive learning: From interactive activation to adaptive resonance. Cognitive Science, 11, 23–63.
HurwitzJ.B. (1991). A hidden-pattern unit network model of category learning. Unpublished manuscript, Brooks AFB, , TX.
14.
KnappA.G.AndersonJ.A. (1984). Theory of categorization based on distributed memory storage. Journal of Experimental Psychology: Learning, Memory, and Cognition, 10, 616–637.
15.
KohonenT. (1984). Self-organization and associative memory. Berlin: Springer-Verlag.
16.
KruschkeJ.K. (1992). ALCOVE: An exemplar-based connectionistic model of category learning. Psychological Review, 99, 22–44.
17.
MackintoshN.J. (1975). A theory of attention: Variations in the associability of stimuli with reinforcement. Psychological Review, 82, 276–298.
18.
MalsburgC. von der. (1973). Self-organization of orientation sensitive cells in the striate cortex. Kybernetik, 14, 85–100.
19.
McClellandJ.L.RumelhartD.E. (1985). Distributed memory and the representation of general and specific information. Journal of Experimental Psychology: General, 114, 159–188.
20.
McCloskeyM. (1991). Networks and theories: The place of connectionism in cognitive science. Psychological Science, 2, 387–395.
21.
MeditchJ.S. (1969). Stochastic optimal linear estimation and control. New York: McGraw-Hill.
22.
MyungI.J. (1989, August). Cue learning in a numerical prediction task. Paper presented at the Twenty-Second Annual Mathematical Psychology Meeting, University of California, Irvine.
23.
MyungI.J.BusemeyerJ.R. (1992). Measurement-free tests of a general state-space model of prototype learning. Journal of Mathematical Psychology, 36, 32–67.
24.
NosofskyR.M.KruschkeJ.K.McKinleyS. (1992). Combining exemplar-based category representations and connectionistic learning rules. Journal of Experimental Psychology: Learning, Memory, and Cognition, 18, 211–233.
RumelhartD.E.HintonG.E.WilliamsR.J. (1986). Learning internal representations by error propagation. InRumelhartD.E.McClellandJ.L., (Eds.), Parallel distributed processing: Explorations in the microstructure of cognition: Vol. 1. Foundations (pp. 318–362). Cambridge, MA: MIT Press.
27.
SearleS.R. (1982). Matrix algebra useful for statisticsNew York: Wiley.
28.
ShanksD.R. (1990). Connectionism and the learning of probabilistic concepts. Quarterly Journal of Experimental Psychology, 42A, 209–237.
29.
ShanksD.R. (1991). Categorization by a connectionistic network. Journal of Experimental Psychology: Learning, Memory, and Cognition, 17, 433–443.
30.
SuttonR.S.BartoA.G. (1981). Toward a modern theory of adaptive networks: Expectation and prediction. Psychological Review, 88, 135–170.
31.
WerbosP. (1974). Beyond regression: New tools for prediction and analysis in the behavioral sciences. Unpublished doctoral dissertation, Harvard University, Cambridge, MA.
32.
WhiteH. (1989a). Learning in artificial neural networks: A statistical perspective. Neural Computation, I, 425–464.
33.
WhiteH. (1989b). Some asymptotic results for learning in single hidden-unit feedforward network models. Journal of the American Statistical Association, 84, 1003–1013.
34.
WidrowB.HoffM.E. (1960). Adaptive switching circuits. Institute of Radio Engineers, Western Electric Show and Convention, Convention Record, 4, 96–104.
