Effects of Averaging Data During Stimulus Generalization

Migler, Bernard

doi:10.1901/jeab.1964.7-303

Cited by 70 publications

(40 citation statements)

References 10 publications

(7 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…It is more likely that the differences are due to the response pacing contingency in the present experiment. Data reported by Blough (1963) and others (Crowley, 1979;Cumming & Eckerman, 1965;Migler, 1964) suggest that continuous generalization gradients may be due to the mixing of discrete patterns of responding with other behaviors to produce intermediate rates of responding. It has been hypothesized that when a temporal property of a response (i.e., IRT, duration) is reinforced differentially in the presence of two All subjects correctly categorized these stimuliaccording to type on first viewing.…”

Section: Resultsmentioning

confidence: 91%

Discrimination and generalization of leaf damage by blue jays (Cyanocitta cristata)

Real

Iannazzi

Kamil

et al. 1984

Animal Learning & Behavior

View full text Add to dashboard Cite

Blue jays (Cyanocitta cristata) responded to projected black-and-white silhouettes of cherry leaves that were either undamaged or were damaged by either cryptic caterpillars that disguise leaf damage due to their feeding or by noncryptic caterpillars that do not disguise leaf damage due to their feeding. Pecks to the key on which the images were projected were reinforced only if interresponse times fell within specified temporal boundaries. These boundaries were different in the presence of the two types of leaf damage. Following training with one exemplar of each damage type, the jays correctly categorized novel instances of both types. They categorized novel undamaged leaves as if they were examples of leaves showing damage by cryptic caterpillars. Results suggest that the jays can readily discriminate and generalize to new members of a polymorphous stimulus class. The results suggest that techniques for the study of concept formation in animals can be applied to an animal's ability to categorize stimuli that are involved in a potential prey-predator relationship.Many palatable, highly cryptic, leaf-eating caterpillars have feeding behaviors that appear to disguise or eliminate evidence of leaf damage due to feeding. These caterpillars pare leaves evenly, often snipping off the leaves on which they have fed. Unpalatable, noncryptic caterpillars lack these behaviors. These caterpillars leave abundant evidence of their feeding. The leaves on which unpalatable caterpillars have fed are often left in tatters (Heinrich, 1979). Heinrich and Collins (1983) have suggested that the feeding stategies of the cryptic caterpillars may have evolved under predation pressures to further reduce their visibility, and they have demonstrated that black-capped chickadees (Parus atricapillus) can use both artificial and naturally occurring leaf damage as a means of localizing prey items. Their data indicate that birds are capable of discriminating leaf damage and that leaf damage may determine the locus of foraging efforts. Their data suggest, more indirectly, that foraging behaviors may be determined, at least in part, by leaf or leaf-damage characteristics as a result of

show abstract

Section: Resultsmentioning

confidence: 91%

Discrimination and generalization of leaf damage by blue jays (Cyanocitta cristata)

Real

Iannazzi

Kamil

et al. 1984

Animal Learning & Behavior

View full text Add to dashboard Cite

show abstract

“…Cumming and Eckerman (1965), Migler (1964), Migler and Millenson (1969), and Wildemann and Holland (1972) each trained animals to produce discrete response patterns in the presence of two stimulus values. Subsequent generalization tests included stimuli intermediate in value to the two training stimuli.…”

mentioning

confidence: 99%

Microanalysis of variable-interval performance during stimulus compounding

Bushnell

Weiss

1978

Animal Learning & Behavior

View full text Add to dashboard Cite

Additive summation is observed when more responses are emitted to the simultaneous presentation (tone-plus-light) of independently conditioned stimuli (tone and light) than to either stimulus presented alone. The current experiment sought to determine if this increased rate during tone-plus-light was a function of a new modal interresponse time (lRT) or a differential mixing of pauses with a modal IRT characteristic of the responding in tone and light alone. Three rats were trained On a three-component multiple schedule where tone and light were each associated with a variable-interval30-sec schedule while a variable-intervallOO-sec operated in the simultaneous absence of these stimuli, tone-offand light-out. Baseline response rates were 2-4times as high in tone or light as in their absence. In testing, more responses were emitted to tone-plus-light than to tone or light by all animals, but the modal IRT was in the O.2-o.4-sec IRT bin for all test conditions. Tone-plus-lightcontrolled fewer long IRT values and more responses in the short modal category than tone or light alone. These results support the response mixing hypothesis of stimulus control; i.e., no "new" behavior was observed during the novel combination of stimulus elements, only a mixture of previously reinforced behavior patterns in different proportions.On a generalization test, stimulus control of behavior can be assayed in molar terms, i.e., as average response rates, and/or in molecular terms, i.e., as interresponse times (IRTs). In molar terms, a systematic relationship exists between response rates to the generalized stimuli and the similarity of these stimuli to the training stimulus (SD), with the peak at the SD value (Guttman & Kalish, 1956). Are these lowered response rates in the presence of the generalized stimuli merely a mixing of previously reinforced behaviors with pauses or are the behavior patterns unique to those stimuli? In order to answer this question, Blough (1963), Crites, Harris, Rosenquist, and Thomas (1967), and Sewell and Kendall (1965 collected IRT data during stimulus generalization tests. Each found that the stimulus generalization gradient was produced by an increase in the number of long IRTs, while the modal IRT value remained unchanged. They concluded that the rates to the generalized stimuli were a product of mixing the short IRT's characteristic of SD responding with long pauses.Other investigators demonstrated the mixing of behavior more explicitly. Cumming and Eckerman

show abstract

“…Skinner's "kinds of evidence" for strength -including whether a response is emitted at all, its energy level, its speed, the amount of repetition, and its overall frequency -suggest that strength is a property underlying observed behavior. Constructions such as "evidence of strength," "sign of strength," and "indicator of strength" occur at least 15 times in Verbal Behavior, making strength come across as a kind of "state variable" which is not directly observable (Shull, 1993) Shimp, 1969), stimulus generalization (e.g., Migler, 1964), and so-called continuous repertoires (Wildemann & Holland, 1972).…”

mentioning

confidence: 99%

The concept of strength in the analysis of verbal behavior

Cherpas

1993

Analysis Verbal Behav

View full text Add to dashboard Cite

A textual analysis of the concept of strength is presented, based on patterns of its occurrence in the book Verbal Behavior (Skinner, 1957). A manual count was conducted of all instances of the word "strength," and closely related forms. Rates were then plotted and interpreted as revealing the kinds of situations where strength is most relevant. Strength appears to be most relevant (as measured by instances per page of text) whenever a detailed behavioral analysis involves 1) more than one source of strength for a response, 2) multiple responses or fragments being strengthened by one or more variables, 3) dynamic changes in behavior, or 4) behavior which is not currently or readily observed. Further research is needed to evaluate how textual analysis of this sort contributes to a science of behavior.

show abstract

Effects of Averaging Data During Stimulus Generalization

Cited by 70 publications

References 10 publications

Discrimination and generalization of leaf damage by blue jays (Cyanocitta cristata)

Discrimination and generalization of leaf damage by blue jays (Cyanocitta cristata)

Microanalysis of variable-interval performance during stimulus compounding

The concept of strength in the analysis of verbal behavior

Contact Info

Product

Resources

About