I am looking for performance results from categorization studies with discrete stimulus features and reinforcement feedback. I need something like this:

stimulus  - correct response - performance after X trials
(1,1,1,0) -         0        -   20.1% correct
(1,1,0,1) -         1        -   23.3% correct
(1,0,0,0) -         1        -   82.0% correct

(e.g. Shepard, Hovland & Jenkins, 1961; Medin & Schaffer, 1978)

Medin, D. L., & Schaffer, M. M. (1978). Context theory of classification learning. Psychological Review; Psychological Review, 85(3), 207.
Shepard, R. N., Hovland, C. I., & Jenkins, H. M. (1961). Learning and memorization of classifications. Psychological Monographs: General and Applied, 75(13), 1-42.

  • 3
    $\begingroup$ can you tell us what you're trying to do? as it is, the information you're requesting is very broad (who knows how each author defines 'performance'?) and the numbers in one study most likely are not comparable to any other. $\endgroup$
    – Jeff
    Jan 18, 2013 at 16:43
  • 2
    $\begingroup$ Have you already done a forward search and looked at some of the articles that cite these two? $\endgroup$ Jan 19, 2013 at 0:22
  • $\begingroup$ Jeff: i have a computational model i'm trying to validate across as many studies as possible. by 'performance' i simply mean that i want to know how often correct/incorrect responses are given (rather than neural-responses, eye-data, etc). $\endgroup$
    – vdv
    Jan 21, 2013 at 22:32
  • $\begingroup$ Artem: yes, i've done a forward search. the problem is that i get too many studies that do not follow the format described above. the signal-noise ratio is too low to be useful. any help here would be appreciated. $\endgroup$
    – vdv
    Jan 21, 2013 at 22:36
  • $\begingroup$ @vdv, please incorporate any relevant extra information in your question as well. Comments should only be seen as 'temporary'. The feedback you gave can easily be edited into your question. $\endgroup$
    – Steven Jeuris
    Jan 28, 2013 at 10:49

3 Answers 3


You've described a very general type of study design which has probably been employed in hundreds of studies or more. Your question might profit from a more specific description of the type of study you're looking for.

A few examples that fit the bill just off the top of my head are:

Goldstone (1996) Isolated and interrelated concepts

Zaki and Homa (1999) Concepts and transformational knowledge

Kornell and Bjork (2008) Learning concepts and categories: is spacing the "enemy of induction"?

  • $\begingroup$ thank you baixiwei, i have edited my question, making it more specific. i will follow up on those references, but please let me know if the new format of the question brings different papers to mind. $\endgroup$
    – vdv
    Jan 18, 2013 at 15:51

Of course, the type of data you use depends lot on what your model purports to explain. The way i take your question, you are looking for straight up artificial classification learning experiments. Is this correct?

If so, John Kruschke has published several of his most important datasets on his website. I have often used these in my model testing. His '93 and '96 papers in connection science are especially notable.

Unfortunately, Nosofsky has not done the same thing, but you can find PDFs on his website, many of which contain reports of data (i.e., tables) detailed enough to simulate. I'd especially look at his '86 paper in JEP General, but there is tons on his publications page and that's just the start...

Finally, one of my favorite, under-simulated, datasets was published in the '92 cogsci proceedings by Aha & Goldstone. PDF here


Bob Rehder's work may be of interest, for instance:

Rehder, B., & Hoffman, A. B. (2005). Thirty-something categorization results explained: Selective attention, eyetracking, and models of category learning. Journal of Experimental Psychology: Learning, Memory, and Cognition, 31, 811-829.

You will see some eye-tracking data on some classic results, but if you are after the so called performance aspects of the data, this will be reported also.

You might also find the following data interesting, they are working with discrete category dimensions and corrective feedback:

Blair, Mark R., et al. "Extremely selective attention: Eye-tracking studies of the dynamic allocation of attention to stimulus features in categorization." Journal of Experimental Psychology: Learning, Memory, and Cognition 35.5 (2009): 1196.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.