Effect of intertrial interval duration on component processes in concept learning.

Wells, Herbert

doi:10.1037/h0032769

Cited by 2 publications

(8 citation statements)

References 5 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…As with the win-stay assumption, the obtained data show a moderate (Coltheart, 1973a;Erickson, 1968;Gumer & Levine, 1971;Kornreich, 1968;Levine, 1966;Schwartz, 1966) to substantial (Rogers & Haygood, 1968;Suppes & Schlag-Rey, 1965;Wells, 1972) deviation from the "strong" form of the lose-shift assumption.…”

Section: Negative Feedbackmentioning

confidence: 70%

“…Levine (1966) found data support for this idea. Other researchers, however, found this inconsistency to be higher, ranging anywhere from 10% to 26% (Chumbley, 1969;Coltheart, 1971Coltheart, , 1973aKornreich, 1968;Wells, 1972).…”

Section: Response Generationmentioning

confidence: 91%

“…Another problem with the blank-trial manipulation is that the high consistency of blank-trial patterns corresponding to given hypotheses seems to be partly a function of the overtraining that Levine's subjects received. When given only 1 (Coltheart, 1971) or 2 (Wells, 1972) test problems, rather than 16 (Levine, 1966), subjects displayed less consistency. Coltheart found 76% and Wells noted 74% of blank-trial protocols matched a single hypothesis, as compared to 92% that Levine obtained.…”

Section: No Feedbackmentioning

confidence: 95%

“…However, there is a good deal of evidence suggesting that the strong form is untenable. A large number of studies have shown the existence of win-shift behavior (Coltheart, 1973a;Cotton, 1971Cotton, , 1972Dodd & Bourne, 1969;Erickson, 1968;Gumer & Levine, 1971;Karpf & Levine, 1971;Kornreich, 1968;Levine, 1966Levine, , 1971Nahinsky, 1968;Nahinsky & Slaymaker, 1969;Rogers & Haygood, 1968;Schwartz, 1966;Suppes & Schlag-Rey, 1965;Wells, 1972), ranging from as little as 2% (Erickson, 1968) to as much as 50% (Schwartz, 1966) of the time, depending on procedural variations. In addition, Kenoyer (1972) found that the prediction breaks down not only in simple comparisons between two successive trials but also over a series of reinforcements.…”

Section: Positive Feedbackmentioning

confidence: 99%

“…There is some evidence supporting the theoretical viability of the local consistency concept. Wells (1972) discovered that as the intertrial interval in concept identification problems is increased, subjects exhibited local consistency on a greater number of occasions, thus suggesting that they do use such a tactic naturally because its frequency increases as the normal intertrial interval (about 7 seconds) is approached. Also, Erickson (1968) found that a model based on local consistency sampling gave a good fit to his obtained data.…”

Section: Partial Memorymentioning

confidence: 99%

See 4 more Smart Citations

Examination of hypothesis-sampling theory.

Brown¹

1974

Psychological Bulletin

View full text Add to dashboard Cite

Recent work on human concept identification behavior has stressed the role of hypotheses in guiding subjects' responses. This article deals with the empirical validity of the assumptions made by the various models of hypothesis behavior. Specifically, five general classes of assumptions are examined: basic (exclusion of hypothesis learning, hypothesis as the unit of behavior, and response generation); learning (all-or-none versus strength); transition (hypothesis switch after positive, negative, and no feedback); sampling of single hypotheses (assuming no memory, perfect memory, and partial memory); and sampling of groups of hypotheses (subset and dimensional sampling). Certain other methodological and interpretive problems in the literature are also pointed out.

show abstract

Section: Negative Feedbackmentioning

confidence: 70%

Section: Response Generationmentioning

confidence: 91%

Section: No Feedbackmentioning

confidence: 95%

Section: Positive Feedbackmentioning

confidence: 99%

Section: Partial Memorymentioning

confidence: 99%

See 3 more Smart Citations

Examination of hypothesis-sampling theory.

Brown¹

1974

Psychological Bulletin

View full text Add to dashboard Cite

show abstract

Effects of Response Modes on Concept Identification

Forest

Crossman

1976

Psychol Rep

View full text Add to dashboard Cite

40 subjects solved three concept-identification problems under one of four treatment conditions. Purpose of the experiment was to determine how the response format used by subjects affected their performance. Subjects used one of four possible response modes: verbalize a single hypothesis when they thought they had the correct answer, verbalize a new hypothesis each time they decided to switch answers, classify each stimulus as positive or negative, or classify stimuli as positive or negative and verbalize a hypothesis on each error. Preliminary analysis indicated that both main effects, problems and response format, were significant. However, a correction for the implicit criterion used by subjects who verbalized a single hypothesis suggested that these subjects were able to learn as fast as those in the other treatment conditions and that response format did not differentially affect concept-identification performance.

show abstract

Effect of intertrial interval duration on component processes in concept learning.

Cited by 2 publications

References 5 publications

Examination of hypothesis-sampling theory.

Examination of hypothesis-sampling theory.

Effects of Response Modes on Concept Identification

Contact Info

Product

Resources

About