Academia.eduAcademia.edu
All Topics
Psychology
Experimental Psychology
The Experimental Analysis of Behavior

The Experimental Analysis of Behavior

description2,404 papers
group233 followers
lightbulbAbout this topic
The Experimental Analysis of Behavior is a scientific approach within psychology that studies behavior through controlled experiments, focusing on the relationships between environmental variables and behavioral responses. It emphasizes the use of empirical methods to understand the principles of behavior and the effects of reinforcement and punishment.
lightbulbAbout this topic
The Experimental Analysis of Behavior is a scientific approach within psychology that studies behavior through controlled experiments, focusing on the relationships between environmental variables and behavioral responses. It emphasizes the use of empirical methods to understand the principles of behavior and the effects of reinforcement and punishment.
On two types of deviation from the matching law: bias and undermatching1
by William M Baum

1974, Journal of the Experimental Analysis of Behavior

Data on choice generally conform closely to an equation of the form: log(B%/B2) = a log(r1/r2) + log k, where B, and B2 are the frequencies of responding at Alternatives 1 and 2, r, and r2 are the obtained reinforcement from Alternatives... more
descriptionView Paperarrow_downwardDownload
Area under the curve as a measure of discounting
by Joel Myerson

2001, Journal of the Experimental Analysis of Behavior

We describe a novel approach to the measurement of discounting based on calculating the area under the empirical discounting function. This approach avoids some of the problems associated with measures based on estimates of the parameters... more
descriptionView Paperarrow_downwardDownload
Economic concepts for the analysis of behavior
by Steven Hursh

1980, Journal of the Experimental Analysis of Behavior

A review of the relationship between schedule of reinforcement, response rate, and choice suggests that certain unifying concepts from economics can contribute to a more complete science of behavior. Four points are made: 1) a behavioral... more
Fig. 1. Data from two experiments showing overall responses per hour as a function of FR size. Circles (Collier, Hirsch, & Hamlin, 1972) are two rats working 24 hrs per day. Squares (Felton & Lyon, 1966) are from a repre- sentative pigeon studied in short sessions; data were derived by computations from postreinforcement pause and running-rate measures provided in the original report (details on request). Consider an experiment in which animals must emit a certain number of responses to
Fig . 2. Data from two experiments showing overall responses per hour as a function of VI values (sec). Filled circles are from a representative rhesus monkey working for food concurrently with water. The pro- cedure was essentially that described in Hursh (1978) except all food was obtained from a single VI schedule (see discussion of Experiment I). Open circles are from a representative pigeon described by Catania & Reyn- olds (1968).
subjects showed a strongly increasing response rate over the same range of schedules that Ca- tania and Reynolds found decreased response rate. These results are analogous to those in the FR studies; an increase in constraint on reinforcement altered: response rate in conflict- ing ways. Based on this evidence, schedule type, session length, and food deprivation per se can be eliminated as critical factors. Species differences are not important since both mon- keys (Hursh, 1978) and rats (Hursh & Natel- son, in press, see Figure 10 below; see also Figure 5 in Graft, Lea, & Whitworth, 1977) show an increasing function with VI schedules. I propose that in these and all other behav- ioral studies the economic system controlling consumption can strongly determine the re- sults. In the studies by Felton and Lyon (1966) and Catania and Reynolds (1968), the subjects were held at a fixed body weight (80% of ad lib feeding weight) and given supplemental feeding to keep food intake about constant. The total daily consumption of food was not the result of the subjects’ interaction with the environment during the sessions, but was arbitrarily controlled by the experimenter. This is what I call an open economy. In the
Fig. 3. The theoretical basis for behavioral equilib- rium based on the intersection of the subject’s de- mand (d) and the environment’s supply (s). An increase in demand is shown as d’ with the consequent change in equilibrium point.
Fig. 4. Representative data from Hursh (1978), Ex- periment I (filled circles) and Experiment II (open circles), showing response rate as a function of variable- interval schedules. Data were computed from the sum of responses on two concurrent VI schedules for food as a function of the average variable-interval between food pellets from both schedules.
periment, are not identical. The demand curve, as I define it, is not entirely determined by the organism or the reinforcer, but is an out- come of an experiment. Its shape may depend on the method of measurement just as the shape of a generalization gradient depends on the schedules of reinforcement (e.g., Hearst, Koresko, & Poppen, 1964).2 periment, are not identical. The demand curve, as I define it, is not entirely determined by the organism or the reinforcer, but is an out- come of an experiment. Its shape may depend on the method of measurement just as the shape of a generalization gradient depends on the schedules of reinforcement (e.g., Hearst, Koresko, & : Poppen, 1964)? 1
Fig. 6. The two response rate functions derived from the demand curves in Figure 5. Price at each equilib- rium point was multiplied by the quantity consumed and plotted as a function of the schedule of rein- forcement, FR (top panel) or VI (bottom panel).
Fig. 9. Mean response rate by three rats working 24 hrs per day for electrical brain stimulation (EBS) or food on equal VI schedules. Data reported by Hursh and Natelson (in press).
Fig. 10. Definitions of inelastic, unit elastic, elastic and mixed demand in terms of the demand curves, the response rate curves and the elasticity coefficients of the log-log demand curves (see text for explanation). gasoline per capita has changed very little over the last ten years as the price of gasoline has quadrupled. In contrast, attendance at movies is greatly influenced by price. The boundary condition between these two categories is unit demand in which each change in price pro- duces an equi-proportional change in quantity demanded (Watson & Holman, 1977, Chap- ter 3). Response rate is mathematically determined by demand. It is the quantity consumed times the unit price. The relationship between de- mand and response rate is summarized in the center panels of Figure 10. If quantity de- manded decays gradually with increases in price (P) then gross expenditure or response rate (R) will be an increasing function of price. For example, since demand for gasoline is inelastic, per capita expenditure for gaso- line has increased with the increasing prices. If demand is a steeply decaying function, then, by contrast, response rate will be a decreasing function. For example, in some families the demand for fish is elastic. As price for fish has increased, they have nearly stopped eating it and their annual expenditures for fish have
Fig. 12. Demand curves and response output func- tions for food, water, and light by one monkey work- ing 24 hrs per day reported by Findley (1959).
Fig. 14. Pellets consumed of various sizes as a func- tion of fixed ratio by rats working 24 hrs per day re- ported by Hill (Note 2).
Fig. 16. Choices per day for food and heroin infusions as a function of intertrial interval (trials per day) by two baboons working 24 hrs per day. Data reported by Elsmore (Note 1).
Fig. 17. Top panels: demand curves for food pellets as a function of FR and bar-tray distance with 16 hrs (left) or 22 hrs (right) deprivation. Bottom panels: response output functions based on demand curves above (note doubling of y-axis in right panel). Data from Roper (1975).
Fig. 18. Demand curves for food pellets determined with varying VI schedules in a closed economy (closed circles) and an open economy (open circles). Data are from Hursh (1978), Experiments I and II; computa- tions are explained in legend of Figure 8.
Fig. 19. Top panel: Substitution—reductions in de- mand for one food source (d,, de, ds, etc.) as a function of increases in the supply (and reductions in price) of a substitutable food source (8, 16, 29, etc. pellets per hr). Bottom panel: Complementation—increases in demand for water (d,, dz, dy, dy, etc.) as a function of increases in the supply (and reductions in price) of a comple- mentary commodity—food. Data from Hursh (1978). and by reducing the supply of the comple- ment, cheap downtown parking.
Fig. 20. Increase in demand for ethanol when shifted from a food satiated baseline to a food deprived base- line. Data from Meisch and Thompson (1973) replotted in log-log coordinates. ment, Water. 1M UUs case just tle Opposite change in demand occurred; as the supply of food, the complement of water, increased from zero to 58 pellets. per hour, the demand for water moved to the right. The subject paid an increasing price for water to complement the increasing supply of food. The same commod- ity that substituted for food and reduced de- mand, complemented water and increased demand. This is a third case in which two re- sponses under similar schedules of reinforce- ment (but for different commodities) were affected differently by a common variable—in this case, a change in a third schedule of rein- forcement. a . ge a ¢-. a ee . - +
log-log coordinates to permit comparisons of elasticity. They studied rats drinking ethanol under FR supply schedules. When food sa- tiated, the subject’s demand was low as in the filled circles. When access to food was re- stricted, demand for ethanol increased uni- formly. Here, the calories from the ethanol substituted for calories lost from food so that a restriction in food led to an increase in de- mand for the substitute ethanol. Note that elasticity of demand (the slope) in this case was not altered. Generally, in open economies where a supplemental supply is provided, this supply will alter demand in the test session by substitution. This can reduce demand. When the amount of a substitutable supply is varied across conditions to maintain constant weight or intake, there will be an apparent increase in elasticity of demand within the test situation. This explains the decreasing response rate functions found by Felton and Lyon (1966) and Catania and Reynolds (1968) shown in the first two figures. They provided a substitutable source of food between sessions that made demand for food in the sessions elastic.
descriptionView Paperarrow_downwardDownload
Delay or probability discounting in a model of impulsive behavior: effect of alcohol
by Suzanne Mitchell and 
2 more
Harriet Wit
J. Richards

1999, Journal of the Experimental Analysis of Behavior

Little is known about the acute effects of drugs of abuse on impulsivity and self-control. In this study, impulsivity was assessed in humans using a computer task that measured delay and probability discounting. Discounting describes how... more
Fig. 1. Individual discounting curves for the 12 least impulsive participants. For each participant, the delay con- dition data are displayed on the left and the probability condition data are displayed on the right. The solid line represents the best fit of the relevant hyperbolic function (Equation 1 or Equation 3). Panels are arranged according to the values of k, which indexes the steepness of the hyperbolic function in the delay condition. Beginning at the top left, the lst participant (Participant 6) had the smallest k in the delay condition, and at the bottom right, the last participant (Participant 9) had the median k in the delay condition. The values of k and h for each subject are given in Table 1. To facilitate comparisons between the delay and probability condition data, the five probability values (1.0, .9, .75, .5, and .25) are expressed as odds against winning (0, 0.11, 0.33, 1.0, and 3).
Fig. 2. Individual discounting curves for the 12 most impulsive participants. Beginning at the top left, the Ist participant (Participant 13) had the median k in the delay condition, and the last participant (Participant 8) had the greatest k in the delay condition. Other details as for Figure 1.
Fig. 3. Scatter plot of k in the delay condition versus hin the probability condition after natural log transfor- mation, that is, In(k) versus In(h). Each data point rep- resents the value of 4 for probability as a function of the value of k for delay for an individual participant. The k and h values are the hyperbolic model data presented in Table 1.
Fig. 4. Discounting curves for the group data. The data points represent the median amount of immediate money estimated to be equivalent to the delayed $10 (top panel) and to the probabilistic $10 (bottom panel). The solid curve represents the best fit of a hyperbolic dis. counting function (Equations 1 and 3), and the dotted curve represents the best fit of an exponential discount ing function (Equations 2 and 4).
@ The nonlinear curve-fitting program was unable to provide delay 7 values for Participant 6.
Note. DSST = Digit Symbol Substitution Test, DEQ = Drug Effects Questionnaire, ARCI = Addiction Research Center Inventory. * Significantly different from preplacebo or preethanol baseline at p < .05 according to Wilcoxon signed-ranks matched-pairs 1 test. ** Significantly different from preplacebo or preethanol baseline at p < .01 according to Wilcoxon signed-ranks matched-pairs / test.
Pearson product-moment correlation coefficients for es- timates of the k and hf parameters in discounting func- tions and scores of impulsivity in personality measures.
Note. Participants 1, 5, 9, 11, 12, 15, 18, and 19 were female; the remainder were male. Values of k and h for each participant before and after administration of placebo and the low (0.5 g/kg) and moderate (0.8 g/kg) doses of ethanol.
Delay indifference points for each participant before and after administration of placebo and the low (0.5 g/kg) and high (0.8 g/kg) doses of ethanol. The values in the table are in dollars. The delays are in days (2, 30, 180, 365). The zero-delay point is not included because it was always equal to $10.00.
Probability indifference points for each participant before and after administration of placebo and the low (0.5 g/kg) and high (0.8 g/kg) doses of ethanol. The values in the table are in dollars. The 1.0 probability point is not included because it was always equal to $10.00.
Results of ANOVAs of indifference points for each level of delay and probability (i.e., delay, prob factor), pre- versus postbeverage (i.e., drug factor). Eight separate analyses were con ducted, including analyses for placebo and alcohol sessions in the low-dose group (n = 12) and high-dose group (n = 12). Note. In all the F tests the degrees of freedom were (1, 11) for the drug factor, (2, 44) for the delay (or probability) factor, and (4, 44) for the interaction between drug and delay (or probability).
Breath alcohol levels, performance on the Digit Symbol Substitution Test (psychomotor task) and the “feel drug” scale of the Drug Effects Questionnaire (DEQ) for each participant before and after administration of placebo and the low (0.5 g/kg) and high (0.8 g/kg) doses of ethanol.
Scores on the like drug scale of the Drug Effects Questionnaire (DEQ) and the sedation and euphoria scales of the Addiction Research Center Inventory (ARCI) for each participant be- fore and after administration of placebo and the low (0.5 g/kg) and high (0.8 g/kg) dose: of ethanol.
Note. Maximum scores for the various tests: IVE impul- sivity scale = 19, EPI impulsivity scale = 9, EPI extrover- sion scale = 24, and SSS disinhibition scale = 10. Larger values indicate greater impulsivity. Values of various personality measures for each partici- pant. The measures are for the impulsivity scale of the Impulsiveness-Venturesomeness-Empathy Questionnaire (IVE), the impulsivity and extroversion scales of the Eysenck Personality Inventory (EPI), and the disinhibi- tion scale of the Sensation Seeking Scale (SSS).
descriptionView Paperarrow_downwardDownload
Discounting of delayed rewards: Models of individual choice
by Joel Myerson

1995, Journal of the Experimental Analysis of Behavior

descriptionView Paperarrow_downwardDownload
Matching, undermatching, and overmatching in studies of choice
by William M Baum

1979, Journal of the Experimental Analysis of Behavior

Almost all of 103 sets of data from 23 different studies of choice conformed closely to the equation: log (B1/B2) = a log (r1/r2) + log b, where B, and B2 are either numbers of responses or times spent at Alternatives 1 and 2, r, and r2... more
Fig. 1. Appearance of overmatching and undermatch- ing when relative responding and reinforcement are represented as proportions. Different shapes arise from different values of a in Equation 2.
Fig. 2. Proportion of variance accounted for by least- squares regression using Equation | plotted against the fitted slope a. Horizontal lines indicate 80% and 100% of variance accounted for. Note logarithmic x-axis; see text for further explanation. The distributions of Figure 3 show varia- tion to both sides of a single mode. They are Figure 2 indicates that Equation | generally described the data well. The majority of points lie above .9. Out of 103 sets, less than 80% of the variance was accounted for in only eleven sets; less than 70% in only three. The fits to the time ratios and response ratios were equally good. The averages for time and re- sponses were 90.5% and 90.3%. Figure 2 shows also that there was no systematic relation be- tween goodness of fit and slope.
MATCHING, UNDERMATCHING, AND OVERMATCHING Fig. 3. Frequency distributions of the slope a obtained by fitting Equation 1 to the various sets of data. Verti- cal lines show the location of 1.0, the value required by the matching relation. The two distributions sepa- rate studies that measured responses (bottom) and time (top).
A nonparametric test was used to assess sta- tistical significance (Hollander & Wolfe, 1973, pp. 200-204). For this, a nonparametric proce- dure was used to fit a regression line (column vii in Table 1; Hollander & Wolfe, 1973, pp. 205-216). The number of data (column ix) was sometimes less than for the least-square method (column iv) because ordinates with identical abscissae were averaged into a single value. The probability that the slope could equal 1.0 (column x) appears in Figure 5 plotted against the nonparametric-estimated slope. Apart from four or five outliers, the points conform to a roughly symmetrical curve fall- ing away to both sides of a peak at 1.0. A hori- zontal line indicates the level of probability equal to .05. Points lie above this line for slopes from about .60 to 1.56. No points lie below the line in the range between approxi- mately .90 and 1.20. According to this test, then, slopes between .9 and 1.2 approximate 1.0. Outside this range, unsystematic variation presumably determines whether the slope should be considered distinct from 1.0. Although Figures 2 and 5 suggest that the Fig. 5. Probability of a true slope of 1.0 (nonpara- metric test of Hollander & Wolfe, 1973, pp. 201-204) versus the slope estimated by a nonparametric fitting technique (Hollander & Wolfe, 1973, pp. 205-206). The graph includes 103 points. The horizontal line shows probability equal to .05.
Fig. 6. Loss in predictability (proportion of variance accounted for) of the method of least-squares from as- suming the matching relation (slope a = 1.0) to be true. See text for further explanation.
descriptionView Paperarrow_downwardDownload
The correlation-based law of effect1
by William M Baum

1973, Journal of the Experimental Analysis of Behavior

It is commonly understood that the interactions between an organism and its environment constitute a feedback system. This implies that instrumental behavior should be viewed as a continuous exchange between the organism and the... more
descriptionView Paperarrow_downwardDownload
Determination of discount functions in rats with an adjusting-amount procedure
by Suzanne Mitchell and 
1 more
J. Richards

1997, Journal of the Experimental Analysis of Behavior

An adjusting-amount procedure was used to measure discounting of reinforcer value by delay. Eight rats chose between a varying amount of immediate water and a fixed amount of water given after a delay. The amount of immediate water was... more
descriptionView Paperarrow_downwardDownload
A search for symmetry in the conditional discriminations of rhesus monkeys, baboons, and children
by Ronald Lazar

1982, Journal of the Experimental Analysis of Behavior

Procedures for generating arbitrary matching-to-sample performances may generate only conditional discriminations. Rational grounds for this distinction are proposed, based on the properties that any equivalence relation must possess.... more
descriptionView Paperarrow_downwardDownload
Rule-governed behavior and sensitivity to changing consequences of responding
by Robert Zettle and 
1 more
Irwin Rosenfarb

1986, Journal of the Experimental Analysis of Behavior

Humans were presented with a task that required moving a light through a matrix. Button presses could produce light movements according to a multiple fixed-ratio 18/differential-reinforcement-of-low-rate 6-s schedule, with components... more
descriptionView Paperarrow_downwardDownload
Delay Discounting: I'm a k, You're a k
by Amy Odum

2011, Journal of the Experimental Analysis of Behavior

Fig. 1. Indifference points as a function of delay for an individual nonsmoking participant (NS1) and a smoking participant (SM1) for hypothetical money (NS1 & SM1) and hypothetical cigarettes (SM1). These participants were the first smoker and nonsmoker tested in a study by Bickel, Odum, and Madden (1999), who reported median data across groups. See Bickel et al. for details of the procedure. Lines show the best fit of Equation 1 to the data using GraphPad Prism®.
Goodness of fit (R®) for Equation 1 (the hyperbolic model) and Equation 2 (the exponential model) for the data shown in Figure 1.
descriptionView Paperarrow_downwardDownload
An operant approach to rehabilitation medicine: overcoming learned nonuse by shaping
by Louis Burgio

1994, Journal of The Experimental Analysis of Behavior

A new approach to the rehabilitation of movement, based primarily on the principles of operant conditioning, was derived from research with deafferented monkeys. The analysis suggests that a certain proportion of excess motor disability... more
descriptionView Paperarrow_downwardDownload
Time and memory: towards a pacemaker-free theory of interval timing
by John Staddon

1999, Journal of The Experimental Analysis of Behavior

A popular view of interval timing in animals is that it is driven by a discrete pacemaker-accumulator mechanism that yields a linear scale for encoded time. But these mechanisms are fundamentally at odds with the Weber law property of... more
descriptionView Paperarrow_downwardDownload
Stimulus equivalence and arbitrarily applicable relational responding
by Steven Hayes

1991, Journal of the Experimental Analysis of Behavior

Subjects' responses to nonarbitrary stimulus relations of sameness, oppositeness, or difference were brought under contextual control. In the presence of the SAME context, selecting the same comparison as the sample was reinforced. In the... more
ested in aspects of learning common to all peo- ple. When the experiment begins, the screen in front of you will show some geometric figures. There will be either two or three figures at the bottom of the screen. Your task is to choose one of these figures by using the joystick. The joy- stick controls the movement of a box on the screen. Move the joystick until the box is around the figure you want to choose. Then press the button on the joystick. Sometimes there will be two figures in the bottom section of the screen, and at other times there will be three. You make your choice the same way in either case.
Fig. 5. Testing performance of Subject 2 (pretrained with SAME and OPPOSITE). Dashed lines indicate probes that were above 80% correct, where “correct” is defined as selecting the indicated stimulus. Wavy lines indicate probes were below 80% on at least one of the testing blocks recorded in that section. The letters S and O indicate the relational stimulus presented. Numbers in parentheses indicate the specific testing block (these same numbers are used in Table 2 for cross reference) and the percentage correct. See Table 2 for specific comparison stimuli used and training sequences.
Subject 3 (Pretrained Same/Opposite)
Fig. 7. Testing performance of Subject 4 (pretrained with SAME and OPPOSITE). Dashed lines indicate probes that were above 80% correct, where “correct” is defined as selecting the indicated stimulus. Wavy lines indicate probes were below 80% on at least one of the testing blocks recorded in that section. The letters S and O indicate the relational stimulus presented. Numbers in parentheses indicate the specific testing block (these same numbers are used in Table 4 for cross reference) and the percentage correct. See Table 4 for specific comparison stimuli used and training sequences.
Subject 6 - No Pretraining of Relational Stimuli Fig. 9. Testing performance of Subject 6 (no pretraining on relational stimuli). Dashed lines indicate probes that were above 80% correct, where “correct” is defined as selecting the indicated stimulus. Wavy lines indicate probes were below 80% on at least one of the testing blocks recorded in that section. The letters S and O indicate the relational stimulus presented. Numbers in parentheses indicate the specific testing block (these same numbers are used in Table 6 for cross reference) and the percentage correct. See Table 6 for specific comparison stimuli used and training sequences.
Fig. 10. Testing performance of Subject 8 (pretrained with SAME and DIFFERENT). Dashed lines indicate probes that were above 80% correct, where “correct” is defined as selecting the indicated stimulus. Wavy lines indicate probes were below 80% on at least one of the testing blocks recorded in that section. The letters S and D indicate the relational stimulus presented. Numbers in parentheses indicate the specific testing block (these same numbers are used in Table 7 for cross reference) and the percentage correct. See Table 7 for specific comparison stimuli used and training sequences. Subject 8 (Pretrained Same/Different)
Fig. 11. Testing performance of Subject 9 (pretrained with SAME and DIFFERENT) in Experiment 1 (top half) and her performance in Experiment 2 after being pretrained with SAME, DIFFERENT, and OPPOSITE (bottom half). Dashed lines indicate probes that were above 80% correct, where “correct” is defined as selecting the
Fig. 12. Testing performance of Subject 7 (pretrained with SAME and DIFFERENT). Dashed lines indicate probes that were above 80% correct, where “correct” is defined as selecting the indicated stimulus. Wavy lines indicate probes were below 80% on at least one of the testing blocks recorded in that section. The letters S and D indicate the relational stimulus presented. Numbers in parentheses indicate the specific testing block (these same numbers are used in Table 9 for cross reference) and the percentage correct. See Table 9 for specific comparison stimuli used and training sequences.
Subject 8 (Pretrained Same/Opposite/Different) Fig. 15a. Testing performance of Subject 8 (pretrained with SAME, DIFFERENT, and OPPOSITE) in Ex- periment 2. Dashed lines indicate probes that were above 80% correct, where “correct” is defined as selecting the indicated stimulus. Wavy lines indicate probes were below 80% on at least one of the testing blocks recorded in that section. The letters S, O, and D indicate the particular relational stimulus presented. Numbers in parentheses indicate the specific testing block (these same numbers are used in Table 10 for cross reference) and the percentage correct. See Table 10 for specific comparison stimuli used and training sequences.
Percentage of correct responses on training problems and probes for Subject 1 (SAME/OP. POSITE pretrained). The criterion for mastery was 90% accuracy for the block of probes and no lower than 80% accuracy on any given type of probe. Usually. when a subject failed to achieve these accuracy rates, he or she was given a review of all pre- viously trained problems; this was followed by a return to the same block of probes where the difficulty was encountered. Because repeated reviews might provide the subject with feed- back about inaccuracy, if three reviews did not produce accurate responding on a given probe block, other problems were presented before returning. to the problematic set of probes. : a er . i a oe RESULTS Subjects Who Receiwed Same/Opposite Pretraining AMES ee eee Se ee Le ee a Use of expanded probe sets. Sometimes a sub- ject failed to show correct responding even af- ter reviewing trained relations. Because we had chosen to use an abbreviated set of all possible probes, it then was possible to give subjects additional related probes without pro- viding any feedback about current or previous responses. Several studies have shown that testing of relations alters the emergence of other relations (e.g., Harrison & Green, 1990; Ken- nedy & Laitinen, 1988). The use of expanded probe sets contributed to the complexity of the resulting designs for individual subjects. A thorough description of the exact training and testing sequences actually used in a given case is more easily understood in the context of the Results section and is presented there.
Percentage of correct responses on training problems and probes for Subject 2 (SAME/OP- POSITE pretrained).
Percent of correct responses on training problems and probes for Subject 3 (SAME/OPPOSITE pretrained). The roles of specific arbitrary visual stimuli (Figure 2) were randomly reassigned for Sub- ject 4’s training and probes (e.g., the stimulus that functioned as A1 for earlier subjects might now be the B2 stimulus). This was done to make sure that some incidental feature of the stimuli had not produced the pattern of control observed with the first 3 subjects. Subject 4 mastered the trained A-B relations quickly and then demonstrated B-A mutual entailment and combinatorial entailment (see Figure 7 and Table 4). After A-C training, Subject 4 failed to pick B2 in probe S[C2]B1-B2 (Test 5 in Figure 7). A scheduled session break occurred at that point, so in the next experimental ses- sion previously trained relations were re- viewed. Subject 4 responded correctly to all probes for mutual entailment and combina- torial entailment of the A, B, and C stimuli. Following C-D training, Subject 4 responded incorrectly to probes for D-B relations (Test 12 in Figure 7). After probes for the inter- mediate A-D relations, his D-B performances rose to 88% correct, and after another test of
Percentage of correct responses on training problems and probes for Subject 5 (no pretraining)
Percentage of correct responses on training problems and probes for Subject 9, Experiment 1 (SAME/DIFFER- ENT pretraining). ence of SAME, showing B-A and C-A mutual entailment and B-C combinatorial entailment. A1, B2, and C2 also entered into an equiva- lence class, but in the presence of OPPOSITE. For example, given O[C2]B1-B2, these sub- jects chose B2. They had learned to pick both B2 and C2 given A1 in the presence of OP- POSITE, and thus B2 and C2 were in an equivalence class given OPPOSITE. AY 2f abi ewes fee A whi nem. mer ews C12
* Change of problems normally used in training or testing. Percentage of correct responses on training problems and probes for Subject 7 (SAME/DIF FERENT pretraining).
Percentage of correct responses to training and probe trials for Subject 8, Experiment 2 (SAME/ DIFFERENT/OPPOSITE pretraining).
2 Change of problems normally used in testing.
2 Change of problems normally used in testing. Percentage of correct responses on training and probes for Subject 9, Experiment 2 (SAME/ DIFFERENT/OPPOSITE pretraining).
descriptionView Paperarrow_downwardDownload
A cost-benefit analysis of demand for food
by Steven Hursh

1988, Journal of the Experimental Analysis of Behavior

descriptionView Paperarrow_downwardDownload
Negative reinforcement as shock-frequency reduction1
by Philip Hineline

1966, Journal of the Experimental Analysis of Behavior

Fig. 1. Schema of the experimental procedure. The tape advances at regular 2-sec intervals. Deflections on the lines marked “post-shock” and ‘‘post-response” indicate holes in these respective channels of the punched tape. The “distribution in effect” line shows which of these channels controls the delivery of shock. Responses and shocks are shown as deflections on the lines indicated. The delivery of shock coincides with the occurrence of a hole in the tape channel currently in control. Control is changed from one channel to the other by a shock if a response has occurred since the last shock, and by a response if a shock has occurred since the last response. Figures 2 and 3 show the course of acquisi- tion for three naive rats, littermates M-1, M-2, and M-3, with parameters .3-.1. Other rats did not appear to differ consistently from these. Figure 2, showing total responses per minute as a function of successive sessions, reveals a variety of acquisition curves. M-1 pressed the lever 130 times during the first 100-min ses- sion, but then dropped to a low rate which increased only gradually, leveling off at about the 38th session; M-2 emitted 60 responses ‘Throughout each session a multi-channel paper tape was advanced one step every 2 sec without exception. Only two channels were used, each corresponding to one of the two random sequences of shock, and at any given time only one of the channels could program a shock. If the channel in control turned up with a punched hole, the brief shock was de- livered; if there was no hole, there would be no shock. The frequency and pattern of holes in each channel were taken from a random- number table; hence the holes turned up ir- regularly but with a specified probability. One channel was in control until a response transferred it to the other channel. Control re- verted to the original channel with the next shock. The two channels are referred to as the post-shock distribution and the post-response distribution, in accordance with the events that put them in control. A single response transferred control to the post-response distri- bution until the next shock, and additional
Fig. 3. Common logarithm of the latency (in sec- onds) between shocks and responses for three rats, M-1, M-2, and M-3, during the first 30 sessions of condi- tioning with shock probabilities equal to .3-.1. Each latency was measured from the shock which followed the previous response. Each point is a mean taken over a 100-min session.
Fig. 4. Showing the correlation between effective (post-shock) response rates and ineffective (post-re- sponse) response rates during avoidance acquisition. Each point is the mean response rate for one rat, taken over five sessions beginning with the first session in which avoidance responding became apparent.
Fig. 7. Distribution of responses in successive 10ths of the 2-sec tape-advancing program for three rats, M-1, M-2, and M-3, with shock probabilities of .3-.1. The ordinate gives the mean number of responses per 100- min session in each .2-sec class interval. The end of the 10th class interval is contiguous in time with the begin- ning of the first class interval, when the probability distributions were sampled by the programming ap- paratus.
Fig. 8. Response rate as a function of shock proba- bilities (and equivalent shock rates). The open circles show rates obtained when the post-shock probability was constant at .3, and the post-response probabilities varied as indicated on the abscissa. The filled circles represent response rates obtained when the post-re- sponse probability was constant at .1, and the post- shock probability varied as indicated on the abscissa. Each point is a mean of the performances of at least two rats, and, in most cases, of four or five rats. The mean rates for individual rats were taken over the final seven days of stable performance on each set of probabilities. which may prove to be based on a logarithmic relation between rate of responding and mag- nitude of reduction in shock probability.
Fig. 9. Cumulative response curve for rat M-1 show- ing conditioning, extinction and reconditioning of avoidance. The vertical dotted lines mark transitions from one pair of shock probabilities (post-shock and post-response) to another, the new set of values being given to the right of each dotted line. Altogether, nine extinction curves, after various parameter values, were obtained. The unmistakable conclusion to be drawn from these is that extinction is slower the closer the parameter values are to one another, which is to say that the resistance to extinction is in- versely related to the pre-extinction rate of re- sponding (given in Fig. 8) for this procedure. Figure 10 shows the average number of presses in extinction as a function of the difference between the post-shock and _post-response probabilities during the preceding condition. The n’s in the figure refer to the number of ex- tinctions, which is equal to the number of rats except for the .4 value where the number of rats was only two. avoidance behavior. It presents in cumulative form the experimental history of a rat origi- nally trained with parameters .3-.1 and extin- guished with parameters .]-.1. Total, cumu- lated presses are plotted against cumulated time under experimentation. After a gradual acquisition period lasting about 3500 min (z.e., 35 sessions), the rat responded at a fairly uni- form rate for another 3000 min before being submitted to extinction. During about 17,000 min (170 sessions) of extinction, the rat emit- ted about 20,000 responses at a decreasing rate. Eventually lever pressing virtually stopped, but was readily restored by the original pa- rameters, .3-.1.
descriptionView Paperarrow_downwardDownload
Effects of increment size and reinforcer volume on progressive ratio performance1
by William Hodos

1963, Journal of the Experimental Analysis of Behavior

The progressive ratio schedule requires the subject to emit an increasing number of responses for each successive reinforcement. Eventually, the response requirement becomes so large that the subject fails to respond for a period of 15... more
descriptionView Paperarrow_downwardDownload
The transfer of respondent eliciting and extinction functions through stimulus equivalence classes
by Michael Markham

1994, Journal of The Experimental Analysis of Behavior

Two studies investigated the transfer of respondent elicitation through equivalence classes. In Experiment 1, match-to-sample procedures were used to teach 8 subjects two four-member equivalence classes. One member of one class was then... more
Fig. 2. Schematic overview of procedural phases for Experiment 1. The study consisted of three phases (see Fig- ure 2). The first involved the training of two four-member stimulus equivalence classes. The second involved “on baseline” discriminated classical conditioning, with one symbol from the first equivalence class serving as a positive conditional stimulus (CS+) and one symbol from the second class serving as a negative conditional stimulus (CS—). Electric shock served as the unconditional stimulus (US). The classical conditioning procedure was presented while the subject was engaged in an operant key-press task. The operant task was used for two reasons. First, it served as a way to keep the subject’s attention focused on the computer screen on which the CSs were presented. Sec- ond, we originally intended to use disruption of operant responding by the CS (conditioned emotional responding, e.g., Lyon, 1968) as a measure of classical conditioning. These data, however, proved to be very unreliable and are not reported. cn ob ee es If... ,..... fe
Fig. 3. Data for Subjects 1, 2, 3, 6, 7, and 8 from Phase 1 symmetry and equivalence tests. Percentage correct ove1 successive blocks of testing.
Fig. 4. Data for Subjects 4 and 5 from Phase 1 symmetry and equivalence tests. Percentage correct over successive blocks of testing.
Fig. 5. Peak skin-conductance levels for the B1 probe trial and during the test for transfer to all other stimuli for all subjects.
1966; Rosen & Leitenberg, 1982; Stampfl & Levis, 1967). However, these intervention: typically result in a reduction in responding to a set of stimuli, not just to the specific stimul targeted in the intervention. A fundamenta question here is why the reduction in feat spreads to stimuli that are not specifically in. cluded in treatment and that bear no physica or formal similarity to the target stimulus. One possibility is that the set of fear-eliciting stim- uli are members of an equivalence class, anc extinction to one member of the class transfer: to the other members. Experiment 2 investi: gated this possibility. In order to test for the transfer of extinction it was decided to condition all of the members of an equivalence class, present one member in extinction, and then test for the transfer of extinction to the other members. In an attempt to replicate Experiment 1, after the test for transfer of extinction, one member of the class used in the original conditioning procedure was selected for reconditioning. Subsequently, the transfer of the reconditioning effects to the other members of the class was examined. Be- cause the transfer of respondent elicitation to the Class 1 stimuli was examined subsequent to the initial conditioning and transfer of ex- tinction procedures, the present study is not a direct replication of Experiment 1. Neverthe-
Subject and Review Committee. As in Experi- ment 1, each subject received one to three 200-ms sample shocks to determine an ac- ceptable level of shock for each subject. No subject selected a shock level below 1.0 mA (see Table 3).
Fig. 9. Skin-conductance levels during conditioning, the test for transfer of extinction, and the test for transfer of respondent elicitation for Subjects 9 through 12.
Fig. 10. Skin-conductance responses during conditioning, the test for transfer of extinction, and the test for transfe of respondent elicitation for Subjects 9 through 12.
Shock levels selected by each subject for use in conditioning during Experiment 1.
Trial types used in training and testing two four-member equivalence classes. The operant task was a key press for which points were awarded on a fixed ratio (FR) 250 reinforcement schedule. The points were tal- lied on the computer monitor and were visible to the subject throughout the task. A 5-min baseline on the operant task was taken before classical conditioning began. Table 2 For classical conditioning, B1 served as the CS+ and B2 served as the CS—. Stimulus duration varied randomly between 20 and 40 s to minimize temporal conditioning (Sachs & May, 1969). A delayed conditioning procedure (Kamin, 1965) was used, in which B1 ter- minated with the onset of the shock. The stim- uli were presented in a semirandom order, with the constraint that no more than two B1 or B2 presentations could occur sequentially. The interstimulus interval varied from 90 to 150 s to minimize temporal conditioning ef- fects, and the stimuli were always presented within 10 s of point delivery via the operant task. There was a total of six B1 and six B2
Shock levels selected by each subject for use in conditioning during Experiment 2. less, the transfer of respondent elicitation is assessed. Table 3
descriptionView Paperarrow_downwardDownload
Changing delay discounting in the light of the competing neurobehavioral decision systems theory: a review
by Mikhail Koffarnus and 
1 more
David Jarmolowicz

2013, Journal of the experimental analysis of behavior

Excessively devaluing delayed reinforcers co-occurs with a wide variety of clinical conditions such as drug dependence, obesity, and excessive gambling. If excessive delay discounting is a trans-disease process that underlies the choice... more
descriptionView Paperarrow_downwardDownload
Discounting of delayed food rewards in pigeons and rats: is there a magnitude effect?
by John Slevin

2004, Journal of the Experimental Analysis of Behavior

Temporal discounting refers to the decrease in the present, subjective value of a reward as the time to its receipt increases. Results from humans have shown that a hyperbola-like function describes the form of the discounting function... more
descriptionView Paperarrow_downwardDownload
Functional classes and equivalence relations
by russell maguire

1989, Journal of the Experimental Analysis of Behavior

Three adult subjects were taught a set of two-choice simultaneous discriminations, with three positive and three negative stimuli; all possible combinations of positive and negative stimuli yielded nine different pairs. The... more
descriptionView Paperarrow_downwardDownload
Effects of alternative reinforcement: does the source matter?1
by William M Baum

1972, Journal of the Experimental Analysis of Behavior

In a chamber with a single response key, pigeon's key pecks were reinforced with food according to a variable-interval schedule. In addition, extra reinforcements occurred concurrently according to an independent schedule. In one... more
descriptionView Paperarrow_downwardDownload
The relation between the generalized matching law and signal-detection theory
by Don Tustin

1978

The generalized matching law can be applied to a signal-detection matrix to give two equations. The first relates responding in the presence of the stimulus to the reinforcements for the responses, and the second relates responding in the... more
Fig. 1. The matrix of stimulus and response events from a signal-detection experiment. The numbers of events occurring in each cell are shown by the letters w, X, y, and z.
Analysis of data reported by Green and Swets (1966) and by Stubbs (1976) according to Equations 4 and 5, Equation 10 and Equation 11. VAC is the percentage of data variance accounted for by the least-squares fitted lines.
descriptionView Paperarrow_downwardDownload
The basis of superstitious behavior: chance contingency, stimulus substitution, or appetitive behavior?
by Gary Lucas

1985, Journal of The Experimental Analysis of Behavior

This research examined three explanations for the "superstitious" behavior of pigeons under frequent fixed-time delivery of food: accidental response-reward contingency, stimulus substitution, and elicited species-typical appetitive... more
descriptionView Paperarrow_downwardDownload
The transfer of specific and general consequential functions through simple and conditional equivalence relations
by Barbara Kohlenberg and 
1 more
Steven Hayes

1991, Journal of the Experimental Analysis of Behavior

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... more
Fig. 2. The performance of Subjects 1 through 6 during training and symmetry, equivalence, and transfer tests in Experiment 1. Each data point represents nine trials for the symmetry and equivalence tests and five trials for the ransfer tests.
Fig. 3. The performance of Subjects 7 through 9 during training and symmetry, equivalence, and transfer tests in Experiment 1. Each data point represents nine trials for the symmetry and equivalence tests and five trials for the transfer tests.
Fig. 5. The performance of Subjects 3, 5, and 8 during training and symmetry, equivalence, and transfer tests in Experiment 3. Each data point represents nine trials for the symmetry and equivalence tests and 15 trials for the transfer tests.
* Subjects left experiment before finishing. Subject 23 could not complete the retraining and then discontinued participation. Number of sorting tasks reaching criterion (out of 20) in Experiment 4.
descriptionView Paperarrow_downwardDownload
Facilitation of human tobacco self-administration by ethanol: a behavioral analysis
by Roland Griffiths

1976, Journal of the Experimental …

T'he effect of ethaniol oin the cigarette smoking of alcoholic subjects was studied in a resi(leintial laboratory. Durinlg daily 6-hr sessions, cigarettes were obtained either by request to the ward staff or by operation of a lever... more
descriptionView Paperarrow_downwardDownload
The structure of equivalence classes
by Lanny Fields

1987, Journal of the Experimental Analysis of Behavior

The structure of equivalence classes can be completely described by four parameters: class size, number of nodes, the distribution of "singles" among nodes, and directionality of training. Class size refers to the number of stimuli in a... more
descriptionView Paperarrow_downwardDownload
The economics of daily consumption controlling food-and water-reinforced responding
by Steven Hursh

1978

In the first experiment, two rhesus monkeys earned their entire ration of food and water during daily sessions with no provisions to ensure constant daily intakes. Two variableinterval schedules of food presentations were concurrent with... more
descriptionView Paperarrow_downwardDownload
Normalized demand for drugs and other reinforcers
by Steven Hursh

1995, Journal of the Experimental Analysis of Behavior

The concepts of behavioral economics have proven to be useful for understanding the environmental control of overall levels of responding for a variety of commodities, including reinforcement by drug self-administration. These general... more
descriptionView Paperarrow_downwardDownload
Melioration, matching, and maximization
by William Vaughan

1981, Journal of the Experimental Analysis of Behavior

HARVARD UNIVERSITY Pigeons were studied in an experiment involving two concurrently available response keys. Conditions were such that in the first condition the predictions of melioration (Herrnstein & Vaughan, 1980), minimization of... more
descriptionView Paperarrow_downwardDownload
Choice in a variable environment: every reinforcer counts
by Michael Davison and 
1 more
William M Baum

2000, Journal of the Experimental Analysis of Behavior

Six pigeons were trained in sessions composed of seven components, each arranged with a different concurrent-schedule reinforcer ratio. These components occurred in an irregular order with equal frequency, separated by 10-s blackouts. No... more
Fig. 1. Bird 91. Log response ratios following each successive reinforcer delivered in each of the seven components in each condition of the experiment. Each panel shows data from a particular reinforcer ratio (indicated at the tor of each panel). The different curves show data from conditions that arranged different numbers of reinforcers pe1 component (R/C) and reinforcer rates (R/M) as indicated in the legend. Response numbers for each alternative were summed across all 35 sessions. On the x axis, the value of 0 indicates response ratios prior to the first reinforce: delivery.
Fig. 2. Bird 92. Log response ratios as a function of successive reinforcers delivered in each of the seven com ponents in each condition of the experiment. See Figure 1.
Fig. 3. Bird 93. Log response ratios as a function of successive reinforcers delivered in each of the seven com- ponents in each condition of the experiment. One data point for the 27:1 reinforcer-ratio component fell off the graph. See Figure 1.
Fig. 4. Bird 94. Log response ratios as a function of successive reinforcers delivered in each of the seven con ponents in each condition of the experiment. See Figure 1.
Fig. 5. Bird 95. Log response ratios as a function of successive reinforcers delivered in each of the seven com- ponents in each condition of the experiment. Two data points for the 1:27 reinforcer-ratio component fell off the graph. See Figure 1.
Fig. 6. Bird 96. Log response ratios as a function of successive reinforcers delivered in each of the seven con ponents in each condition of the experiment. See Figure 1.
Fig. 7. Group log response ratios obtained by summing response numbers over the 6 subjects as a function of successive reinforcers delivered in each of the seven components in each condition of the experiment. See Figure 1.
Fig. 8. Group log response ratios in Conditions 7 (12 reinforcers per component, 2.22 reinforcers per minute) and 8 (12 reinforcers per component, six reinforcers per minute) as a function of the arranged log reinforcer ratio in the components for the data collected before the first reinforcer (R1), between the first and second reinforcers (R2), up to the 12th reinforcer. Filled circles show data from Condition 7, and open circles show data from Condition 8.
Fig. 10. Time-allocation sensitivity to reinforcement (from fits exemplified in Figure 8) according to Equation 1 for the group average data (times summed over all 6 subjects) as a function of successive reinforcers delivered in all conditions of the experiment. R/C indicates the number of reinforcers per component, and Hi indicates the conditions in which six (rather than 2.22) reinforcers per minute were arranged. Fig. 9. Response-allocation sensitivity to reinforce- ment (from fits exemplified in Figure 8) according to Equation 1 for the group average data (response num- bers summed over all 6 subjects) as a function of succes- sive reinforcers delivered in all conditions of the experi- ment. R/C indicates the number of reinforcers per component, and Hi indicates the conditions in which six (rather than 2.22) reinforcers per minute were arranged.
Fig. 11. Condition 7 (2.22 reinforcers per minute). Log response ratios in 1:1 and 27:1 reinforcer-ratio com- ponents as a function of successive reinforcers delivered according to the reinforcer ratio in the previous com- ponent, as shown on the legend. The results from the first component of a session, which did not follow any other component, are labeled “first component.”
Fig. 12. Condition 8 (six reinforcers per minute). Log response ratios in 1:1 and 27:1 reinforcer-ratio com: ponents as a function of successive reinforcers delivered according to the reinforcer ratio in the previous com- ponent, as shown on the legend. The results from the first component of a session, which did not follow any other component, are labeled ‘“‘first component.”
Fig. 13. Scatterplots of the relation between log response ratios emitted before the first reinforcer in a component as a function of the log response ratio emitted before the final reinforcer of the previous component. The lines, and associated equations shown on the graphs, show the best fitting linear regressions through the data. The data are from conditions that arranged 2.22 reinforcers per minute.
Fig. 14. Scatterplots of the relation between log response ratios emitted before the first reinforcer in a component as a function of the log response ratio emitted before the final reinforcer of the previous component. The lines, and associated equations shown on the graphs, show the best fitting linear regressions through the data. The data are from conditions that arranged six reinforcers per minute.
Fig. 15. Log response ratios following each reinforcer delivery in sequences of left- and right-key reinforcer deliveries for all conditions of the experiment. For clarity, only the data up to the delivery of the third reinforcer are shown. Dotted lines join points in which right reinforcers were subsequently delivered.
Fig. 16. Log response ratios following selected se- quences of left-right reinforcer deliveries in Conditions 7 (2.22 reinforcers per minute) and 8 (six reinforcers per minute) when 12 reinforcers per component were ar- ranged. The sequences were L, LR, LLR, LLLR, etc., and R, RL, RRL, RRRL, etc. Dotted lines join disconfirmations, in which a reinforcer was delivered for a response on an alternative that was different from prior reinforced re- sponses in that component.
Fig. 17. Sensitivity to reinforcement values from multiple linear regressions between log response ratios and arranged log reinforcer ratios (Equation 2) in the previous and the present components for each successive reinforce1 delivery. R/C indicates the number of reinforcers per component, and Hi indicates the conditions in which six reinforcers per minute were arranged.
Fig. 18. Predictions of sensitivity to reinforcement val- ues obtained from a simulation of the model embodied in Equations 3 and 4. The figure parallels Figure 17.
Sequence of experimental conditions, number of rein- forcers per component, the maximum number of rein- forcers per session available, and the overall arranged rate of reinforcers in each condition of the experiment. In Conditions 4, 6, and 9, each component was presented twice per session.
Values of sensitivity (a) and bias (log c) for each experimental condition prior to each suc- cessive reinforcer in the components. The data used were averaged over the 6 subjects (i.e., the response and reinforcer totals were averaged and sensitivity values were determined from those averages).
descriptionView Paperarrow_downwardDownload
A transformation of respondently conditioned stimulus function in accordance with arbitrarily applicable relations
by Bryan Roche

1997, Journal of the Experimental Analysis of Behavior

Adult male subjects saw a sexual film clip paired with a nonsense syllable (C1). Similarly, an emotionally neutral film clip was paired with a second nonsense syllable (C3). Responses to the nonsense syllables were recorded as skin... more
descriptionView Paperarrow_downwardDownload
Discrimination Theory of Rule-Governed Behavior
by Daniel Cerutti

1989, Journal of the Experimental Analysis of Behavior

descriptionView Paperarrow_downwardDownload
A Selectionist Approach to Reinforcement
by John Donahoe

1993, Journal of the experimental …

We describe a principle of reinforcement that draws upon experimental analyses of both behavior and the neurosciences. Some of the implications of this principle for the interpretation of behavior are explored using computer simulations... more
Fig. 1. A biobehavioral approach to the analysis and interpretation of complex behavior. The experimental analysis of behavior, which is concerned with the effects of environmental manipulations on behavior, is supplemented by the experimental analysis of physiology, which is concerned with the effects of intraorganismic manipulations on intraor- ganismic events and behavior. longer time intervals (so-called molar relations or correlations) is the cumulative effect of mo- ment-to-moment relations among environ- mental, behavioral, and reinforcing events (Donahoe et al., 1982; Rescorla & Wagner, 1972). That is, sensitivity to correlation is the emergent product of sensitivity to contiguity. only contact between [the schedule] and the organism occurs at the moment (emphasis added) of reinforcement. ... Under a given schedule of reinforcement, it can be shown that at the moment of reinforcement a given set of stimuli will usually prevail. A schedule is simply a convenient way of arranging this. (pp. 2-3)
Fig. 3. A schematic diagram of the proposed cortical systems of the brain involved in modifying synaptic effi- cacies in sensory and motor cortices. Stimuli from various sensory channels (e.g., Syisuar and Savprrory) activate neu- rons in the primary sensory cortex. Axons from the pri- mary sensory cortex ultimately end in synapses on neurons in the motor association cortex (e.g., A and V) and on polysensory neurons in the sensory association cortex. Ax- ons from polysensory cells ultimately end in synapses on neurons in the motor association cortex (e.g., AV) and also on cells within the hippocampus (HIPP). Hippocampal CA1 neurons project diffusely via multisynaptic pathways to the sensory association cortex. Within the motor asso- ciation cortex, the axons of some neurons ultimately ac- tivate effectors that lead to behavior. Axons from some neurons within the motor association cortex project back to the ventral tegmental area (VTA) by means of the medial forebrain bundle (MFB). VTA neurons are acti- vated by unconditioned reinforcing stimuli and project diffusely to the motor association cortex and to the synapses of CA1 hippocampal neurons. (See the text for the func- tioning of these neural systems in the selection of pathways mediating environment-environment and environment- behavior relations.)
Fig. 4. The architecture of a selection network, an adaptive neural network whose structure simulates the neural systems mediating reinforcement in the brain. Shown are the diffusely projecting lateral hypothalamic (LH) and ventral tegmental area (VTA) systems that select pathways mediating environment-behavior relations and the diffusely projecting hippocampal system that selects pathways mediating environment-environment relations.
Fig. 5. Response-selection component of a selection network showing networks having three input units, three interior units, and two output units. The input units can be activated by either of two environmental stimuli (S, and S,) and the putative reinforcing stimulus (US). Ac- tivation of the output units simulates the occurrence of the operant response (R) and the unconditioned response (UR) or, when the UR unit is activated by a stimulus other than the US, the conditioned response (CR). Activation of the UR/CR unit also engages the diffuse reinforcing system that modifies connection weights between all coactive units. The upper panel illustrates a classical, or respondent, contingency in that the occurrence of the US is dependent on the prior activation of an input unit, S, The lower panel depicts an operant, or instrumental, contingency in that activation of the US unit is dependent on the prior activation of the output unit, R, by the presentation of S,.
Fig. 7. Simulation of blocking in a three-phase classical conditioning experiment. During the first phase, activatior of the US input unit was contingent on the prior activation of the CS1 input unit. During Phase 2, activation of the US unit was contingent on the activation of both the CS1 and CS2 units. During the test phase, only the CS2 uni was activated, with the learning algorithm disengaged for “probe” tests to determine whether CS2 had acquired contro over the CR unit.
Fig. 8. Simulation of the acquisition of an intradimensional operant stimulus discrimination by a selection network. Shown are the activation levels of the operant (R) unit initiated by the positive stimulus (S+) and the negative stimulus (S—). S+ activated two input units (S, and S,); if the R output unit became active, then the US input unit, which functioned as the reinforcer, was activated. S— activated two input units (S, and S,) but, whether or not the R unit became active, the reinforcer was never presented.
Fig.9. Simulations of the strengthening of connections to a polysensory unit within the stimulus-selection component of a selection network. Shown are the activation levels of a polysensory unit as.a function of the number of concurrent activations of two input units, S, and S,. In one simulation, the magnitude of the simulated VAT reinforcing signal was low (US = 0.1). In the other simulation, the magnitude was high (US = 0.9).
descriptionView Paperarrow_downwardDownload
Pigeons' discrimination of paintings by Monet and Picasso
by Masumi Wakita

1995, Journal of the Experimental Analysis of Behavior

Pigeons successfully learned to discriminate color slides of paintings by Monet and Picasso. Following this training, they discriminated novel paintings by Monet and Picasso that had never been presented during the discrimination... more
descriptionView Paperarrow_downwardDownload
Self-control: teaching tolerance for delay in impulsive children
by Beth Sulzer-Azaroff

1988, Journal of the Experimental Analysis of Behavior

We investigated a procedure to increase the selection of larger, more delayed reinforcers (i.e., more advantageous in the long run) over smaller immediate reinforcers, in an effort to increase a key aspect of self-control in children. Six... more
descriptionView Paperarrow_downwardDownload
Stimuli, Reinforcers, And Behavior: An Integration
by Michael Davison

1999, Journal of the Experimental Analysis of Behavior

We propose that a fundamental unit of behavior is the concurrent discriminated operant, and we discuss in detail a quantitative model of the concurrent three-term contingency that is based on the notion that an animal's behavior is... more
descriptionView Paperarrow_downwardDownload
From molecular to molar: a paradigm shift in behavior analysis
by William M Baum

2002, Journal of the Experimental Analysis of Behavior

A paradigm clash is occurring within behavior analysis. In the older paradigm, the molecular view, behavior consists of momentary or discrete responses that constitute instances of classes. Variation in response rate reflects variation in... more
descriptionView Paperarrow_downwardDownload
Self-control in pigeons under the Mischel paradigm
by J. Grosch

1981, Journal of the Experimental Analysis of Behavior

Walter Mischel studied self-control in preschool children in the following manner: if the child waited for an interval to end, he or she received the more preferred of two reinforcers; if the child responded to terminate the interval by... more
in effect. Throughout this experiment, both more and less preferred hoppers were activated during the wait interval. Under a second, or “toy plus FR 20” condition, the rear key was present and illuminated during the wait inter- val, and every 20th peck to the rear key was re- inforced with a pellet. In all other respects, the procedure was identical to the “no toy” procedure. Thus, this procedure assessed self- control when an alternative response and alter- native reinforcer were available during the wait interval. In the third, or “toy, no rein- forcement” procedure, everything was identi- cal to that in the “toy plus FR 20” condition except that pellets were never given contin- gent upon rear keypecks. The rear key was illuminated during the wait interval, pecks could occur to it, but the pecks were never re- inforced. Here we assessed the effects on self- control of an alternative response when that response did not engender extrinsic reinforce- ment. Fig. 2. Percentage of trials pigeons waited and received the more preferred reward as a function of avail- ability of an alternative keypeck response. Under Toy FR 20, every 20th response to the alternative key was reinforced; under Toy No FR, the alternative keypecks were not reinforced; and under No Toy, the alterna- tive key was not present.
Fig. 1. Percentage of trials pigeons waited successfully and received the more preferred of two rewards. In the first and third conditions, rewards were present for view during the wait period; in the second condition, re- wards were not present during the wait period.
Figure 3 shows time waited by each of the four birds under the two conditions—hopper: not lighted (points) and hoppers illuminated during the wait interval (X’s). Each of the birds waited, on the average, less time when the hoppers were illuminated (overall average of 6.2 sec) than when they were dark (overall average of 10.7 sec). The average wait time: in each of the two conditions for each of the four subjects were used in a paired t-test which showed a statistically significant differ. ence between the two conditions (¢ = 7.947, di = 3, p < .005, 2-tailed). Fig. 3. Amount of time waited without pecking during each trial. On trials shown by dots, the hoppers were dark and inoperative during the wait interval. On trials shown by X’s, hoppers were illuminated during the wait.
interval, and (b) during seven randomly se- lected trials the pigeon was permitted to eat the more or less preferred reinforcer prior to the beginning of the wait interval. This was accomplished by activating both more and less preferred hoppers and their accompanying hopper lights before the start of the wait pe- riod and allowing the pigeon to eat from either of the hoppers for 3 sec. Once the bird had eaten, the hoppers were inactivated and a standard self-control trial followed immedi- ately. Fig. 5. Amount of time during each trial that pigeons waited without pecking. Access to 3 sec of food was given immediately prior to trials indicated by X’s. No food was received prior to trials indicated by dots.
As predicted, prior experience with a more preferred reinforcer for successful waiting led to significantly longer waiting than did prior experience in which, no matter how long the pigeon waited, more preferred reinforcement was impossible. Similiar results were inter- preted by Mischel and Staub (1965) in terms of the expectancies of the children: when they received successful preliminary trials they were more likely to wait for a delayed, more pre- ferred reinforcer than when they had received unsuccessful preliminary training. However, an alternative interpretation of the present Fig. 6. Amount of time waited without pecking as a function of prior experience. In the infinity condition, birds first experienced wait intervals which were infinitely long, i.e., they were never reinforced for waiting; in the 3-sec condition, birds received prior experience with preferred reinforcement for waiting a short, 3-sec inter- val; in the mixed condition, 50% of the prior trials were infinitely long and 50% 3-sec long.
Fig. 7. Percentage of trials pigeons pecked sufficient numbers of times to receive the more preferred of two re- wards. In the first condition, rewards were present for view during the interval; in the second condition, reward: were not present for view. There followed five test sessions for Pigeon 2 and seven sessions for Y14, each consisting of 20 trials: the hoppers were “up” for 10 trials and “down” for the other 10. During the 10 “hoppers up” trials, both hoppers were acti- vated and illuminated for the duration of the 60-sec trial. Subjects were prevented from en- tering the hoppers by the locked hopper doors, as in Experiment 4. During the “hoppers down” trials, the hoppers were dark and in- activated, as is usually the case. Ten hopper up trials were given followed by 10 hopper down trials in one session; in the next session, the order was reversed, and so on.
descriptionView Paperarrow_downwardDownload
Conditional discrimination and equivalence relations: Control by negative stimuli
by Cammarie Johnson

1993, Journal of the Experimental Analysis of Behavior

NEW ENGLAND CENTER FOR AUTISM Three adult subjects were taught the following two-sample, two-comparison conditional discriminations (each sample is shown with its positive and negative comparison, in that order): Al-BlB2, A2-B2B1;... more
descriptionView Paperarrow_downwardDownload
The merger and development of equivalence classes by unreinforced conditional selection of comparison stimuli
by Kimberly Kirby

1988, Journal of the Experimental Analysis of Behavior

Three experiments assessed the likelihood that subjects with histories of equivalence class development would respond conditionally on new discriminations in the absence of differential consequences for responses. In the first two... more
descriptionView Paperarrow_downwardDownload
The development of functional and equivalence classes in high-functioning autistic children: the role of naming
by Svein Eikeseth

1992, Journal of the Experimental Analysis of Behavior

The development of functional and equivalence classes was studied in four high-functioning, preschoolaged autistic children. Initially, all subjects failed to demonstrate match-to-sample relations indicative of stimulus equivalence among... more
descriptionView Paperarrow_downwardDownload
Associative concept learning in animals
by Edward P Wasserman

2014, Journal of the Experimental Analysis of Behavior

Nonhuman animals show evidence for three types of concept learning: perceptual or similarity-based in which objects/stimuli are categorized based on physical similarity; relational in which one object/stimulus is categorized relative to... more
Fig. 1. Transfer of training for individual pigeons demonstrating the class consistent responding developed during many- to-one training (adapted from Urcuioli, Zentall, Jackson-Smith, & Steirn, 1989).
when a delay was inserted between the sample and comparison stimuli. Typically, hue-sample reten- tion functions for pigeons are generally much shallower than line-sample retention functions (i.e., as the delay increases, matching accuracy is better maintained when the samples are hues) (Farthing, Wagner, Gilmour, & Waxman, 1977; Urcuioli et al., 1989; Wasserman, Bhatt, Chatlosh, & Kiedinger, 1987; Zentall, Urcuioli, Jagielo, Jackson-Smith, 1989). However, the correspond- ing hue-sample and line-sample retention func- tions following MTO training are quite similar (i.e., they have comparable slopes) (see Fig. 2; Urcuioli et al., 1989, Experiment 1; Zentall etal., 1989), a result consistent with the hypothesis that samples associated with a common compari- son stimulus are commonly represented. Fig. 2. Retention functions following matching training in which hue samples and line-orientation samples were associated with different comparisons (one-to-one matching) or common comparisons (many-to-one matching) (adapted from Zentall, Urcuioli, Jagielo, & Jackson-Smith, 1989).
Fig. 3. Following many-to-one matching training pigeons were transferred to a successive discrimination in which responding to samples associated with the same comparison stimulus during training were either both reinforced or not reinforced (consistent) or responding to one sample was reinforced but not the other (inconsistent) (adapted from Kaiser, Sherburne, Steirn, & Zentall, 1997). Note. R=red, G=green, V=three vertical black lines, H=three horizontal black lines, C=a line-drawn circle, D=a 2mm dot. All lines and shapes were projected on a black background. For each trial type the first letter represents the sample and the second the correct compari- son. The transfer test for Group Consistent consisted of simple successive discrimination in which the samples presumed to represent one stimulus class were discriminated from the samples presumed to represent the other stimulus class. The transfer test for Group Inconsistent consisted of a similar simple successive discrimination but one sample from each presumed stimulus class had to be discriminated from the other sample in its presumed class.
Fig. 4. Following many-to-one matching training, sample-comparison associations were reversed for only the hue samples, only the line samples, or for all of the samples (adapted from Zentall, Sherburne, Steirn, Randall, Roper, & Urcuioli, 1992).
Fig. 5. Retention functions following many-to-one matching training in which red and food samples were associated with choice of a vertical-line comparison and green and no-food samples were associated with choice of a horizontal-line comparison (adapted from Zentall, Sherburne, & Urcuioli, 1995).
For further information on how to annotate proofs, click on the Help menu to reveal a list of further options:
Note. R=red, G=green, V= three vertical white lines, H=three horizontal white lines, C=a white line-drawn circle, D =a white dot. All lines and shapes were projected on a black background. For each trial type, the first letter represents the sample and the second the correct compari- son. In Test, samples from Original Training that were not presented in Interim training were presented with the comparisons from Interim Training. Delayed matching tests. _ Evidence consistent with the development of common n representations +11 Again, what specific “cue” (cf. Miller & Dollard, 1941) could be the basis of these emergent stimulus relations? Urcuioli et al. (1989), working within the animal memory literature, suggested that pigeons had commonly represented or “coded” (cf. Lawrence, 1963) samples that were associated with the same comparison stimulus. In general terms, MTO training altered the way in which the organisms implicitly responded to those samples. To test this hypothesis, Zentall, Urcuioli, and their associates conducted a series of experiments to obtain convergent evidence regarding the existence and nature of such common representation(s).
Note. R=red, G= green, V= three vertical white lines, H= three horizontal white lines, C=a white line-drawn circle, D =a white dot. All lines and shapes were projected on a black background. For each trial type, the first letter represents the sample and the second the correct comparison. In test, samples from Phase | that were not presented in Phase 2 were presented with the comparisons from Phase 2 and reinforced responding was either consistent with or inconsistent with the presumed common representations. Many-to-one Transfer Design; Interference/Facilitation (after Zentall, Sherburne, & Urcuioli, 1993).
two basic categories into one broader category that was not defined by any obvious perceptual features. In the first step of the experiment, by reinforcing responding to two different keys, pigeons were trained to sort slides into two arbitrary conjoint categories, such as the conjoint category of cars and people and the conjoint category of chairs and flowers (there were 12 exemplars in each of the four basic- level categories). In the second step of the experiment, the pigeons were trained to make a new response to only one stimulus class in each conjoint category. For example, some birds were trained to peck one new key when a car appeared and to peck another new key when a flower appeared. In the third step involving pecks to the two new keys, the pigeons were tested with slides from all four stimulus classes to see whether the conjoint categories that they had learned in the first step would influence their response to the stimulus classes that had been withheld in the second step (see Bhatt & Wasserman, 1989 for related research into the role of perceptual resemblance in similar reassignment training and testing). Would the birds peck the new key to which cars had been reassigned when a person appeared? Would they peck the new key to which flowers had been reassigned when a chair appeared? The average score was 87% correct in the case
Note. R=red, G= green, V=vertical lines, H = horizon- tal lines, C = circle, D = dot. “c” and “d” represent circle and dot mediating responses, respectively, to the samples (R, G, V & H). — represents associations formed during Original Training. => represents associations formed during Interim Training. Thus, in Testing, V and H will elicit mediating responses “c” and “d” and transfer of training should result.
Note. R=red, G=green, V=vertical lines, H = horizon- tal lines, C= circle, D= dot. “r’, “g”, “v’, and “h” represent red, green, vertical, and horizonval mediating responses, respectively, to the samples (C and D). Dashes (-) represent associations formed during Phase 1. Double dashes (=) represent associations formed during Phase 2. Dashes (-) represent associations formed during Phase 1. Double dashes (=) represent associations formed during Phase 2. Stimuli in parentheses represent the actual comparison stimuli in Phase 3. Thus, in Testing, B and W will elicit mediating responses “r” and “g” rather than “v” and “h” and transfer of training should not result.
Note. R=red, G= green, V=vertical lines, H = horizon- tal lines, C= circle, D = dot. “r’, “g”, represent red and green mediating responses, respectively, to the samples (R and G, and V and H). — represents associations formed during original training. => represents associations formed during interim training. Thus, in testing, V and H will elicit mediating responses “r” and “g” and transfer of training should result.
Note. All successive presentations were followed by food. For matching-to-sample the first stimulus represents the sample, the second stimulus the correct comparison. The incorrect comparison is presented in parentheses. All stimuli were counterbalanced.
descriptionView Paperarrow_downwardDownload
The role of multiple-exemplar training and naming in establishing derived equivalence in an infant
by Inmaculada Gómez Becerra

2007, Journal of the Experimental …

The conditions under which symmetry and equivalence relations develop are still controversial. This paper reports three experiments that attempt to analyze the impact of multiple-exemplar training (MET) in receptive symmetry on the... more
descriptionView Paperarrow_downwardDownload
Emergent simple discrimination established by indirect relation to differential consequences
by William V Dube

1988, Journal of the Experimental Analysis of Behavior

Three experiments examined a discrimination training sequence that led to emergent simple discrimination in human subjects. The experiments differed primarily in their subject populations. Normally capable adults served in the first... more
descriptionView Paperarrow_downwardDownload
The Behavioral Pharmacology of Effort-related Choice Behavior: Dopamine, Adenosine and Beyond
by Merce Correa and 
1 more
John Salamone

2012, Journal of the Experimental Analysis of Behavior

For many years, it has been suggested that drugs that interfere with dopamine (DA) transmission alter the ''rewarding'' impact of primary reinforcers such as food. Research and theory related to the functions of mesolimbic DA are... more
descriptionView Paperarrow_downwardDownload
A Review of Domestic Dogs' (Canis Familiaris) HumanLike Behaviors: Or Why Behavior Analysts Should Stop Worrying and Love Their Dogs
by Monique Udell

2008, Journal of The Experimental Analysis of Behavior

Dogs likely were the first animals to be domesticated and as such have shared a common environment with humans for over ten thousand years. Only recently, however, has this species' behavior been subject to scientific scrutiny. Most of... more
descriptionView Paperarrow_downwardDownload

Related Topics

  • Cognitive Science
  • Psychology
  • Comparative Cognition
  • Applied Behavioral Science
  • Behavioral Ecology
  • Ancient Philosophy
  • Behavioral Economics
  • Classics
  • Behavior Analysis
  • Bioinformatics
    • Download research papers for free!
    580 California St., Suite 400
    San Francisco, CA, 94104