Stimulus generalization of behavioral history

Undergraduates responded under a variable-ratio 30 schedule in the presence of a 25-mm long line and on a differential-reinforcement-of-low-rate 6-s schedule when a 13-mm long line was present. Following this, a line-length continuum generalization test was administered under a fixed-interval 6-s schedule (Experiment 1) or extinction (Experiment 2). In both experiments, obtained generalization gradients conformed to typical postdiscrimination gradients. Responses were frequent under stimuli physically similar to the 25-mm line and infrequent under stimuli physically similar to the 13-mm line. The generalization gradients were generally asymmetric with peak response rates occurring at line lengths greater than 25 mm.     

Peak shift revisited: a test of alternative interpretations

In Experiment 1, 2 groups of human subjects were trained to respond to 1 of 2 light intensity stimuli, S2 or S4, and then were tested for generalization with a randomized series of increasing values from S1 to S11. Both groups, including the group trained to respond to dimmer value, showed peak shifts to a brighter more centrally located test stimulus. In Experiment 2, which used line angle stimuli, both the size of the difference between S+ and S- and the range of test stimuli that extended beyond S+ were varied. The larger the S(+)-S- separation and the larger the range, the greater was the peak shift obtained. In Experiment 3, training involved an S- (line angle) surrounded by 2 S+ values with testing symmetrical about the training values and covering either a narrow or a wide range. The wide range produced greater peak shifts in both directions from S-. All 3 experiments support an adaptation-level interpretation of intradimensional discrimination learning and generalization test performance in human subjects. Related work with animals suggests the presence of similar processes.     

The transfer of specific and general consequential functions through simple and conditional equivalence relations

The purpose of this study was to examine the transfer of consequential (reinforcement and punishment) functions through equivalence relations. In Experiment 1, 9 subjects acquired three three-member equivalence classes through matching-to-sample training using arbitrary visual forms. Comparison stimuli were then given conditioned reinforcement or punishment functions by pairing them with verbal feedback during a sorting task. For 8 of the 9 subjects, trained consequential functions transferred through their respective equivalence classes without additional training. In Experiment 2, transfer of function was initially tested before equivalence testing per se. Three of 4 subjects showed the transfer without a formal equivalence test. In Experiment 3, 3 subjects were given training that gave rise to six new three-member conditional equivalence classes. For 2 of the subjects, the same stimulus could have either a reinforcement or punishment function on the basis of contextual cues that defined its class membership. Experiment 4 assessed whether equivalence training had established general or specific consequential functions primarily by adding novel stimuli in the transfer test. Subjects treated even novel feedback stimuli in the transfer test as consequences, but the direction of consequential effects depended upon the transfer of specific consequential functions through equivalence relations.     

Sequential effects in dimensional contrast

Two experiments examined pigeons' response rates during short trials signaled by stimuli closely spaced along a wavelength continuum. In Experiment 1 separate halves of the continuum were correlated with different reinforcement schedules. In Experiment 2, the middle stimulus was accompanied by a lower probability of reinforcement than were the remaining wavelengths. Both procedures resulted in dimensional contrast “shoulders,” seen as relatively enhanced or depressed response rates in the presence of stimuli between the extreme of the continuum and the border separating the positive and negative stimuli. Sequential analyses addressed possible contributions of the following interactions: (a) local contrast, seen when rate during a given schedule depends on the schedule in the just-preceding trial; (b) modification of local contrast by the similarity of the signaling stimuli (P. Blough, 1983); and (c) schedule-independent rate contrast, seen when rate in a given trial depends on the rate controlled by the stimulus that accompanied the just-preceding trial (Malone & Rowe, 1981). Dimensional contrast functions were similar when isolated according to the schedule, to the similarity of the signaling stimulus, and to the response rate of the just-preceding trial. The interactions noted above do not appear to make important contributions to this effect.     

Stimulus duration as a measure of stimulus generalization

Four pigeons in the line-positive group were trained with a vertical line on a green background that signalled intermittent reinforcement while a plain green field signalled extinction. Four pigeons in the line-negative group were trained with the opposite discrimination. Response to a control key terminated any trial and initiated the next trial. The birds also used the control key during generalization tests to control the durations of trials in which various line orientations were presented. These durations were summed to provide generalization gradients of stimulus duration that were positive or negative in accordance with the trained discriminations. In Experiment 2, birds from the line-positive group were tested with a procedure in which the control key was not available on some trials. This provided an independent assessment of response rates to the test stimuli. These rates were used to predict the stimulus durations obtained when the control key was available. The findings supported a general model for the prediction of response distributions among concurrent stimuli from rates observed with single stimuli.     

Stimulus generalization of behavioral history: Interspecies generality and persistence of generalization gradients

Four pigeons were exposed to a tandem variable-interval (VI) fixed-ratio (FR) schedule in the presence of a 50-pixel (about 15 mm) square or an 80-pixel (about 24 mm) square and to a tandem VI differential-reinforcement-of-low-rate (DRL) schedule when a second 80-pixel or 50-pixel square was present. The values of the VI and FR schedules were adjusted to equate reinforcement rates in the two tandem schedules. Following this, a square-size continuum generalization test was administered under a fixed-interval (FI) schedule or extinction. In the first testing session, response frequency was a graded function of the similarity of the test stimuli to the training stimuli for all pigeons. These systematic generalization gradients persisted longer under the FI schedule than under extinction.     

Stimulus generalization of gravity

In two experiments, squirrel monkeys were exposed to centrifugally generated, artificial gravity and trained to respond for food reinforcement at selected gravity (g) levels. Experiment I involved a single g value; in Exp. II, subjects were trained to discriminate among two or three g values. After training, generalization tests were administered over a 1.1-g to 2.1-g range. The major findings were as follows: (a) single-stimulus training yielded a linear relationship between percentage of responding and magnitude of gravity. (b) Two-valued discrimination training produced gradient peaks which were shifted from S(D) in a direction away from S(Delta). This effect was cancelled when S(D) was located equidistant between two S(Delta) stimuli. (c) Gradient form was independent of the S(D)-S(Delta) difference, but related to continuum location and/or intensity of discriminative stimuli.     

Inhibitory stimulus control following errorless discrimination learning

Three generalization procedures were used to investigate inhibitory stimulus control following discrimination learning with few errors. Three groups of pigeons acquired a discrimination between a green stimulus (the positive stimulus) and a vertical or horizontal line (the negative stimulus) through differential autoshaping followed by multiple schedule presentation of the two stimuli with gradually increasing stimulus durations. Genereralization testing was along a line-tilt continuum. For one group, the test involved a resistance-to-reinforcement procedure in which responses to all line tilts were reinforced on a variable-interval schedule. For a second group, also tested with the resistance-to-reinforcement procedure, the lines were superimposed on the green field that formerly served as the positive stimulus. A third group was tested in extinction with the combined stimuli. Control groups had no discrimination training but responding to green was nondifferentially reinforced. The control subjects responded more to all line tilts during testing than did the comparable experimental subjects, indicating that the negative stimulus had become an inhibitory stimulus. Both resistance-to-reinforcement groups revealed inhibitory gradients around the negative stimulus, but the gradient for the extinction group was relatively flat. These data are consistent with others that modify Terrace's early conclusion concerning the failure of inhibition to develop during errorless training.     

Peak shift but not range effects in recognition of faces

University students were trained to discriminate between two gray-scale images of faces that varied along a continuum from a unique face to an average face created by morphing. Following training, participants were tested without feedback for their ability to recognize the positive face (S+) within a range of faces along the continuum. In Experiments 1 and 4, the range of stimuli presented during testing was manipulated. In Experiment 2, participants viewed different ranges of faces during an adaptation period that followed training and preceded testing. In all experiments, generalization functions revealed peak shifts or area shifts (fewer “yes” responses to novel faces on the negative side of the S+), but no systematic effects of the test or adaptation range. Peak shift was found both for upright and inverted faces and occurred even if the orientation of the face was reversed between training and test. Using similar methods, either an area shift or range effect (but not both together) was demonstrated for line tilt stimuli (Experiment 3), and the appearance of these effects depended on instructions. It appears that peak shift and area shift are robust across many different kinds of stimuli, but range effects may not readily occur with complex multidimensional stimuli.     

Effects of variable-interval value and amount of training on stimulus generalization.

In Experiment 1 pigeons pecked a key that was illuminated with a 501-nm light and obtained food by doing so according to a variable-interval (VI) schedule of reinforcement, the mean value of which differed across groups: either 30 s, 120 s, or 240 s. The pigeons in all three groups were trained for 10 50-min sessions. Generalization testing was conducted in extinction with different wavelengths of light. Absolute and relative generalization gradients were similar in shape for the three groups. Experiment 2 was a systematic replication of Experiment 1 using line orientation as the stimulus dimension and a mean VI value of either 30 s or 240 s. Again, gradients of generalization were similar for the two groups. In Experiment 3 pigeons pecked a key that was illuminated with a 501-nm light and obtained food reinforcers according to either a VI 30-s or a 240-s schedule. Training continued until response rates stabilized (> 30 sessions). For subjects trained with the 30-s schedule, generalization gradients were virtually identical regardless of whether training was for 10 sessions (Experiment 1) or until response rates stabilized. For subjects trained with the VI 240-s schedule, absolute generalization gradients for subjects trained to stability were displaced upward relative to gradients for subjects trained for only 10 sessions (Experiment 1), and relative generalization gradients were slightly flatter. These results indicate that the shape of a generalization gradient does not necessarily depend on the rate of reinforcement during 10-session single-stimulus training but that the effects of prolonged training on stimulus generalization may be schedule dependent.     

Coding responses and the generalization of matching to sample in children.

Two experiments studied the conditions of stimulus control necessary for the generalization of relational matching to sample. Matching required the selection of comparison shapes rotated 90 degrees clockwise from the orientation of the corresponding sample. In Experiment 1, five children were taught to: (a) code the orientations of samples, (b) transform sample codings to account for the 90 degree rotation, and (c) repeat the transformed sample coding response to a comparison. High levels of generalization occurred with a set of novel stimuli for which stable sample-coding responses were initially available. In another novel set, where stable sample-coding responses were not initially available, low levels of generalized matching were recorded. Matching performance improved after stable coding responses were trained. In Experiment 2, two children and three adults were trained in a form of the matching task that produced poor generalization despite the presence of stable sample-coding responses. Retraining to modify the stimulus control exerted by these coding responses produced an immediate improvement in generalized matching to sample. Results suggest that the generalization of matching is dependent on structure of stimulus control that the component responses exert on each other.     

On the effectiveness of including meaningful pictures in the formation of equivalence classes

In three experiments, 165 adult participants were trained on 12 baseline conditional discriminations and tested for the formation of three 5-member equivalence classes (A➔B➔C➔D➔E). All experiments included two reference groups; the abstract (ABS) group was trained with all abstract stimuli and the picture (PIC) group with C-stimuli as meaningful pictorial stimuli but A, B, D, and E stimuli as abstract shapes. In Experiment 1, the color of the meaningful stimuli was manipulated. In the ABS, PIC, and black-and-white groups, 33.3%, 80%, and 93.3% formed equivalence classes, respectively. In Experiment 2, participants were exposed to a test block with and without trials that included C stimuli. For the groups with and without C trials in the test, 93.3% and 86.7% formed equivalence classes, respectively, compared to 20% in the ABS group. In Experiment 3, the number of meaningful pictures and their location in stimulus classes were manipulated. One group was trained with 3 pictures (C1/B2/D3, the 3-PIC) while the other groups had 2 pictures (C1/B2 and C1/D3, the 2-PIC). In the second test block for the ABS and PIC groups, 6.7% and 86% of the participants formed equivalence classes, respectively. For the 3-PIC and the 2-PIC groups, 66.7% and 50% of the participants formed equivalence classes, respectively. Results suggest that the effects of meaningful stimuli in equivalence classes (a) cannot be attributed to the use of colorful stimuli in previous studies, (b) occur during training and are not dependent on the presence of meaningful stimuli at test, and (c) are sensitive to stimulus location.     

STIMULUS GENERALIZATION AND EQUIVALENCE CLASSES: A MODEL FOR NATURAL CATEGORIES

Two three-member classes were formed by training AB and BC using a conditional discrimination procedure. The A and B stimuli were nonsense syllables, and the C stimuli were sets of “short” or “long” lines. To test for equivalence, C1 or C2 was presented as a sample with A1 and A2 as comparisons. Once the class-related comparison was chosen consistently, different line lengths were substituted for the training lines in the CA tests. In general, the likelihood of choosing a given comparison was an inverse function of the difference in the length of the test line from the training line. Stimuli in an equivalence class became functionally related not only to each other but also to novel stimuli that resembled a member of the equivalence class. The combination of primary generalization and equivalence class formation, then, can serve as a model to account for the development of naturally occurring categories.     

The impact of arbitrarily applicable relational responding on evaluative learning about hypothetical money and shock outcomes

Evaluative learning comprises changes in preferences after co-occurrences between conditioned stimuli (CSs) and an unconditioned stimulus (US) of affective value. Co-occurrences may involve relational responding. Two experiments examined the impact of arbitrary relational responding on evaluative preferences for hypothetical money and shock outcomes. In Experiment 1, participants were trained to make arbitrary relational responses by placing CSs of the same size but different colours into boxes and were then instructed that these CSs represented different intensities of hypothetical USs (money or shock). Liking ratings of the CSs were altered in accordance with the underlying bigger/smaller than relations. A reversal of preference was also observed: the CS associated with the smallest hypothetical shock was rated more positively than the CS associated with the smallest amount of hypothetical money. In Experiment 2, procedures from Relational Frame Theory (RFT) established a relational network of more than/less than relations consisting of five CSs (A-B-C-D-E). Overall, evaluative preferences were altered, but not reversed, depending on (a) how stimuli had been related to one another during the learning phase and (b) whether those stimuli referred to money or shocks. The contribution of RFT to evaluative learning research is discussed.     

On Stimulus Generalization and Stimulus Classes

Stimulus generalization has been defined as the spread of effect of reinforcement for responses emitted in the presence of one stimulus to different stimuli presented under extinction conditions. As a result of stimulus generalization, novel stimuli come to exert stimulus control over members of the response class. Studies in the applied behavior analysis literature, however, have reported experimental preparations that included prompting and reinforcement procedures during what were claimed to be stimulus generalization conditions. These studies violated the procedural requirement that stimulus generalization be tested under extinction conditions. Responses that come under the control of a class of stimuli may do so by direct training or by stimulus generalization. It is desirable for organisms to respond in the presence of members of an appropriately constructed stimulus class, but we should understand the mechanism of entry into the class by its members. If inaccurate claims of stimulus generalization are made when training procedures are used in the ostensible generalization conditions, the robustness of the original training procedures will be over estimated. By adhering to the operational requirements of behavioral definitions, we could better understand the power and limits of our educational and training procedures.     

Toward establishing a qualifying autoclitic repertoire in children with autism spectrum disorder

Autoclitics are secondary verbal operants that are controlled by a feature of the conditions that occasion or evoke a primary verbal operant such as a tact or mand. Qualifying autoclitics extend, negate, or assert a speaker's primary verbal response and modify the intensity or direction of the listener's behavior. Howard and Rice (1988) established autoclitics that indicated weak stimulus control (e.g., “like a [primary tact]”) with four neurotypical preschool children. However, generalization to newly acquired tacts was limited. In Experiment 1, we addressed similar behavior as in Howard and Rice but with autistic children while using simultaneous teaching procedures, and we observed generalization across sets and with newly acquired tacts. In Experiment 2, we evaluated the effects of multiple-exemplar training on generalization of autoclitics across sets of naturalistic stimuli. Across participants, gradual increases in the frequency of autoclitics occurred with untaught stimuli after teaching with one or more sets.     

Visual discrimination, categorical identification, and categorical rating in brief displays of curved lines: implications for discrete encoding processes

Visual discrimination, categorical identification, and categorical rating measurements were made on sets of curved-line stimuli drawn from a theoretically uniform continuum with curvature parameter s. In Experiment 1, discriminability of pairs of curved lines separated by a constant distance on the s scale was measured at successive points along the scale. Curved lines were presented four at a time in a 100-msec display, which was followed by a random-dot mask. Discrimination performance was found to vary nonsmoothly with s. In Experiment 2, a categorical identification task was performed in which subjects labeled the curved-line stimuli of Experiment 1 straight, just curved, and more than just curved. From these data, a theoretical discrimination performance was computed that was closely congruent to the discrimination performance of Experiment 1. In Experiment 3, three different categorical rating scales with two, three, and four intervals were tested and each was shown to be less effective than the categorical identification scale for predicting discrimination performance. Mean ratings were, however, highly linear with s, suggesting that the curved-line continuum was psychometrically uniform. Experiment 4 provided further evidence for the uniformity of the curved-line continuum by measuring conventional acuity for curvature. Two rather than four curved lines were presented in each display; duration was increased to 2 sec; and the poststimulus mask was omitted. Acuity was found to vary linearly with s. It was concluded that under conditions in which attention is distributed over a number of elements in the field and in which viewing and effective visual processing time are restricted, performance in discriminating curved-line stimuli may be determined by relatively coarse, discrete visual processes.     

Response patterning during stimulus generalization in the rat

Nine rats were trained to bar press in the presence of a clicking sound of 6.67 cps (S(D)) for 1-min variable-interval food reinforcement randomly alternated with a clicking sound of 20 cps (S(Delta)) signifying extinction. After a criterion of 90% of total responses in the presence of the S(D) was obtained, a generalization test was administered, including values of 6.67, 10.00, 13.33, and 20.00 cps, with responses in the presence of the S(D) continuing to be reinforced during testing. The test yielded a gradient of response strength with rate highest in the presence of the S(D) and decreasing with increasing distance from this value. An interresponse time (IRT) analysis of responding during generalization testing revealed no systematic differences in modal IRT category or in median IRT to the different test stimuli. Mean IRT was lowest in the presence of the S(D) and increased systematically with increasing distance from this value, supporting the hypothesis that the generalization gradient of response rate is primarily the result of an increasing proportion of "long" IRT responses to stimuli increasingly distant from the S(D).     

A comparison of multiple versus single examples of the correct stimulus on task acquisition and generalization by persons with developmental disabilities

In teaching discriminations to persons with retardation, we often presume we will improve acquisition and generalization if we use multiple examples of boththe correct and incorrect stimuli. Two experiments were conducted to test this hypothesis. In the first experiment, 7 persons with moderate retardation learned to discriminate between functional words under two conditions. In one condition, Multiple Example of S- Only,1 example of the correct stimulus (S+) and 10 examples of the incorrect stimulus (S-) were used during acquisition. In the other condition, Multiple Examples of S+ and S-,10 examples of the S+ and 10 examples of the S- were used. Results showed that the condition which presented only a single example of S+ was superior 16 times and inferior 4 times during acquisition, generalization, and maintenance. A second experiment was conducted to (a) replicate the methodology and procedures in Experiment 1 with different participants, (b) determine whether the results were replicable, and (c) obtain efficiency data. Results replicated the findings of the first experiment. The condition which presented only a single example of S+ was superior on measures of (a) trials to criterion, (b) percent correct during acquisition, and (c) minutes to criterion. On measures of generalization, the two conditions were relatively equal. Thus, the condition which presented only a single example of the correct stimulus was more efficient and was just as effective in generalization as the condition which presented multiple examples of both the S+ and S-. These surprising results were discussed in terms of stimulus control, why students performed just as well during generalization when only one example of the S+ was used, why acquisition was also poorer for this condition, and how future studies might address these points.     

Predicting the extension of equivalence classes from primary generalization gradients: the merger of equivalence classes and perceptual classes.

In Experiment 1, 6 college students were given generalization tests using 25 line lengths as samples with a long line, a short line, and a “neither” option as comparisons. The neither option was to be used if a sample did not go with the other comparisons. Then, four-member equivalence classes were formed. Class 1 included three nonsense words and the short line. Class 2 included three other nonsense words and the long line. After repeating the generalization test for line length, additional tests were conducted using members of the equivalence classes (i.e., nonsense words and lines) as comparisons and intermediate-length lines as samples. All Class 2 comparisons were selected in the presence of the test lines that also evoked the selection of the long line in the generalization test that had been given before equivalence class formation. Class 1 yielded complementary findings. Thus, the preclass primary generalization gradient predicted which test lines acted as members of each equivalence class. Regardless of using comparisons that were nonsense words or lines, the post-class-formation gradients overlapped, showing the substitutability of class members. Experiment 2 assessed the discriminability of the intermediate-length test lines from the Class 1 (shortest) and Class 2 (longest) lines. The test lines that functioned as members of an equivalence class were discriminable from the line that was a member of the same class by training. Thus, these test lines also acted as members of a dimensionally defined class of “long” or “short” lines. Extension of an equivalence class, then, involved its merger with a dimensionally defined class, which converted a close-ended class to an open-ended class. These data suggest a means of predicting class membership in naturally occurring categories.