UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Observing behaviour in pigeons : an investigation of the reinforcing value of the negative discriminitive… Spencer, John Wayne 1974

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata


831-UBC_1974_A8 S64.pdf [ 1.64MB ]
JSON: 831-1.0100008.json
JSON-LD: 831-1.0100008-ld.json
RDF/XML (Pretty): 831-1.0100008-rdf.xml
RDF/JSON: 831-1.0100008-rdf.json
Turtle: 831-1.0100008-turtle.txt
N-Triples: 831-1.0100008-rdf-ntriples.txt
Original Record: 831-1.0100008-source.json
Full Text

Full Text

OBSERVING BEHAVIOUR IN PIGEONS; AN INVESTIGATION OF THE REINFORCING VALUE OF THE NEGATIVE DISCRIMINITIVE STIMULUS by JOHN WAYNE SPENCER B.A., University of California, Davis, 1971 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF ARTS in the Department of Psychology We accept this thesis as conforming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA July, 197^ In presenting th i s thesis in par t i a l fu l f i lment of the requirements for an advanced degree at the Univers i ty of B r i t i s h Columbia, I agree that the L ibrary shal l make it f ree ly ava i lab le for reference and study. I fur ther agree that permission for extensive copying of th is thesis for scho la r l y purposes may be granted by the Head of my Department or by his representat ives. It is understood that copying or pub l i ca t ion of th is thesis for f inanc ia l gain sha l l not be allowed without my writ ten permission. Department of Psychology The Univers i ty of B r i t i s h Columbia Vancouver 8, Canada Date 26 July, 19?k ABSTRACT To determine i f a stimulus associated with nonreinforcement can maintain observing behaviour as predicted by Hendry's informa-tion hypothesis. three pigeons were trained i n a three-key operant chamber. On the yellow center key, periods during which food reinforce-ment for key pecking was available on a random-interval schedule* alternated irregularly with periods of nonreinforcement (extinction). Pecking the yellow right key producedj on a random-interval schedule, stimuli associated with the prevailing reinforcement condition on the center key: green (S+), when the random-interval schedule was in effect; red (S-), when extinction was in effect. Peeking # the yellow left key was reinforced by food on a random-interval schedule. The inclusion of this key was intended to attenuate a possible punishing effect S- might have upon observing by-arranging that reinforcement occasionally occurred on this key when reinforcement for pecking the center key was unavailable. After training, the consequences of peck-ing the observing key were manipulated. In one condition, S+was elim-inated as a consequence and only S- could be produced. In another con-dition, S- was eliminated and only S+ could be produced. When the conse-quences of observing were only S+ or both S+ and S-, observing was main-tained. When only S- was a consequence, observing extinguished. There was l i t t l e difference between the observing rates when only S+ and when both S+ and S- were consequences. The results do not support Hendry»s information theory of observing, but are consistent with reinforcement accounts of observing. i TABLE OF CONTENTS Pag© I Introduction . . . . . . • • • 1 1. Observing behaviour . . . . . . . . . . 1 2. What reinforces observing behaviour? . . . . . . 2 3. Purpose of present research 10 II -i. Method .... . . . . 12 1. Subjects . . 12 2. Appartus . . . . . . . . . . . . . . . . . . . . 12 3. Procedure . . . . . 12 III Results . . . 16 XV Discussion • 20 V References . . . • • . . • • • • • • • • » . • • • . • 2k LIST OF TABLES AND FIGURES Page Figure 1. Wyckoff's assumed relationship between conditioned reinforcing value of a stimulus and frequency of rein-forcement in its presence . . . . . . . 3 Figure 2. Schaub's two-factor theory of the rein-forcing value of S+ and S- . . 6 Table 1. Summary of procedure 14 Table 2. Data summed across last 5 days of each ; condition for each bird . . . . . . . . . . . . . 1? Figure 3. Last 5 sessions of responding on the observing key during the six treatments . . . . . 19 i l l ACKNOWLEDGEMENTS The author wishes to express his abundant gratitude to Dr. D.M. Wilkie, not only for his sublime patience and protracted good-humour, but also for his assistance and expertise in directing this project to its ineffable conclusion. The author is also indebted to Dr. R. Wong for his advice and comments. A special thanks i s due the author1 collegues, who sustained in him a suitable frame of mind for this project by frequently accompanying him to local establishments for therapeutic refreshment. iv INTRODUCTION 1. Observing Behaviour Responses that result in exposure to discrirainitive stimuli that are reliably correlated with different schedules of reinforcement with* out affecting reinforcement probability have been called observing  responses (Wyckoff. 1952; 1959; 1969). In the fi r s t studyken observing behaviour, Wyckoff (1952? 19^ 9) equipped an experimental chamber for pigeons with two operanda. One of these was: acircular, response key, which, when pecked at the appropriate time, produced food reinforcement. The other was a foot treadle situated below the response.key. The key was normally illuminated with white light. Periods of reinforcement for key pecking, arranged according to a 30-s®eond fixed-interval sched-ule, alternated in a quasi-random sequence- with periods of nonreinforce-ment (extinction) of the same duration. Sinw both schedules were associated with the same stimulus, Wyckoff*s pigeons were responding on a mixed reinforcement schedule (Mix FT 30-see Ext). At any time during a session, depression of the foot treadle produced one of a pair of dlscrlminitive stimuli (S+ or S-) on the response-key-by switch-ing the white light to either red or green. The key became red (S+) i f the FT 30-second schedule was in effect, or green (S-) i f extinction was programmed. In this manner, thetreadle observing jresponsessignalled the current reinforcement schedule, and changed the mixed-schedule to a multiple schedule (Mult FT 30-sec Ext). Wyckoff (1952; 1969) found that the pigeons consistently, depressed the foot treadle when the red and green lights were reliably correlated 1 2 with the two schedules. When red and green were uncorrelated with the two schedules, however, the pigeons did not make the observing response. 2, What Reinforces Observing Behaviour? The interpretation of Wyckoff*s (1952; 1969) finding has evolved as the principal issue of observing behaviour. Since observing responses do not increase the frequency of food reinforcement, the question of what reinforces these responses arises. One interpretation Is a condi-tioned reinforcement hypothesis. Wyckoff (1952; 1959) believed that i t was the conditioned reinforcing value of the observing response-produced stimuli that maintained the observing response. He assumed:^  that the con-ditioned reinforcing value of these stimuli was 6^te«nlned~by^e.yproba.-bl l i t y or frequency of reinforcement in their presence. When white was present, periods of reinforcement (FT) and of extinction were not sig-nalled. Hence, the frequency of reinforcement in white was the mean value of the summed frequencies of both PI (120 reinforoements per hour) and Ext (0 reinforcements per hour), i.e.,. 60 reinforcements per hour. When red and green signalled the periods of FT and Ext, the frequency of reinforcement was either aero, in the case ofthe stimulus associated with Ext, or 120 reinforcements per hour, in the case of the-stimulus associated with the FT 30-seeond schedule. The pigeons preference for red and green (the multiple schedule) over white (the mixed schedule) was assumed by Wyckoff to be due to a higher mean conditioned ^ inforcing value for red and green than for white. Wyckoff assumedrthat:the func-tion relating conditioned reinforcing value to probability or frequency of reinforcement was positively accelerated. This relationship is i l l u s -trated in Figure 1. A REINFORCEMENTS/HOUR Figure 1: Wyckof^s (1959) assumed relationship between conditioned reinforcing value of a stimulus and frequency of reinforcement in its presence (after Hendry, I969). 4. A second hypothesis, proposed by Berlyne (195?; I960), maintained that on a schedule each as Mix FT Ext, there is an element of uncer-tainty arising from the unslgnalled alternation between- periods of FT and Ext. This "uncertainty" produces a state of conflict over whether "to peck or not to peck." Since i t can be presumed that this uncer-tainty and the conflict i t induces is aversive, the reduction of uncer-tainty made possible by the outcome of the observing response is the source of reinforcement for the observing response* Hendry (1969) elaborated the proposal-of Berlyne into an informa-tion hypothesis which incorporated the uncertainty aspect but eschewed the notion of conflict. Hendry proposed that the information about rein-forcement or nonreinforcement produced by the observing response is in itself reinforcing. In other words, i t i s the "knowledge gained" of the current circumstance which reinforces the observing response. Accor-dingly, the S- should be equally as effective as the S+ in reinforcing observing behaviour. Thisprediction represents a major difference between the two hypotheses; Wyckoff^ (1952; 1959) hypothesistpredicts that only S+ can reinforce observing behaviour, since only S+ has con-ditioned reinforcing value. A third hypothesis, which is a combination of the Wyckoff and Hendry notions, has been advancedby Schaub (1969). Schaub, while assuming that the Information provided by S+ and S~ i s reinforcing, also acknowledged that associative factors arising from a contempora-neous relationship between the stimulus and primary reinforcement also affect the final value of S+ and S-. Schaub proposed a two-factor model. Reward value derived from information provided by S+, and S- i s equal; the association of S+ with reinforcement provides reward value for S+, 5 while S- acquires an aversive value from its association with nonrein-f orcement. These two factors summate.to determine the final value of S+ and S-f as shown in Figure 2^  Evidence often regarded as supporting Hendry* s(1969) interpre-tation has come from experiments by Schaub (1969) and by Lieberman (1972), which purport to show that S- can reinforce observing behaviour. In Schaub's study, only one response key was available, on which pecking produced food reinforcement. In addition,every third peck on the nor-mally white key changed the colour to red or green for a brief5 period, depending upon whether a variable-interval (VI) period or an;extinction period was currently in effects In the experiment from which Schaub (1969) concluded that S- was reinforcing, he addedobserving: response-independent presentations of the discriminitive stimuli—S+when the VI schedule was in effect, S- when extinction was in effect. Since this procedure signalled the two schedule components, the observing response-produced stimuli, which provided no new Information, were therefore redun-dant. Under this procedure, Schaub found that the observing response rate during both the VI and the Ext periods underwent a sharp decline when compared to the baseline procedure during which the observing response-produced stimuli were not redundant. Since the reinforcing properties of both S+ and S- were eliminated (in the redundant situ-ation), Schaub deduced that the S- must have been reinforcing during the baseline condition* Schaub»s (I969) results* however, have been widely criticized on procedural grounds (see Dinsmoor, Browne and Lawrence, 1972; Jenkins and.... Boakes, 1973; Mulvaney, Jwaideh, Smith and Hughes, 197^ » for details). As a result, his findings are, at best, tenuous. Figure 2: Schaub's (I969) two-factor theory of the reinforcing value of S+ and S-, The broken line shows the net value of the two factors (after Hendry, I969). 7. Lieberman's (1972) monkeys were trained to depress an observing lever which produced 6 seconds of tone (S+) when a VI schedule was in effect on the food lever* or 6 seconds of light (S-) when extinction was scheduled. When the light was eliminated as a consequence of obser-ving, the observing rate during the extinction component underwent a sharp deoline, a result Lieberman attributed to the elimination of the reinforcing properties of the S-. But Dinsmoor et al . (1972) have crit-icized Lieberman^ measure of observing response rates, noting that the situation without the light is not comparable to the situation; with the light. When an observing respo ving responses were recorded until i t went out. When the light was dele-ted as an outcome of observing, a continuous rate of responding was pos-sible throughout the Ext period. In an effort to insure comparable ditlons, Lieberman (1972) did not record a portion of-the responses during that part of the experiment when the light was absent. Under this arrange-ment, he reported a lower observing rate without the light. Dinsmoor et a l . (1972) have argued that Lieberman»s selective recording of responses was not an equitable means of comparing the two conditions* - Some, of^t^ corded responses when the light was not available may have been those that would otherwise have been postponed until the light went out in the condition when the light was available as an outcome of observing. This would account for a higher observing rate in the presence of S- (the light) than in its absence. Other experimenters have reported no decline in observing response rate when the S- was eliminated under.similar circumstances (Schaub, 1969). The equivocal nature of these findings point up some of the dif-ficulties in devising a decisive test of the two hypotheses. Studies 8. by Kendall and Gibson (1965)» Dinsmoor, Flint, Viemeister.and Smith (1969), Dinsmoor et al, (1972), Lieberman (1972), and Mulvaney et a l . (1974) have a l l utilized some form of the stimulus-deletion procedure, which involves establishing an observing response with two disoriraini-tive stimuli (S+ and S-), followed by the removal of one, so that the observing response will produce only one of the schedule-correlated stimuli at the appropriate time. It has been assumed that the stimulus which remains as a consequence of an observing response s t i l l provides information about the schedule component with which i t is associated, while an observing response made during the schedule component for which the associated stimulus has been deleted has been assumed to result in no information. Wilton and Clements (1971a; 1971b), however, have pointed out a difficulty in interpreting results obtained from this type of procedure, since an observing response which does not produce a discriminitive stimulus (e.g., S+) can be just as Informative as the presence of the other (S-). In order to accurately assess the reinforcing properties of S+ or S- alone, this confounding of variables must be overcome. Dinsmoor et al . (1972) have reported a procedure which circum-vented this difficulty by the use of a random-interval (RI) observing response requirement. On a RI schedule, an observing response that does not produce a discriminitive stimulus may indicate only that the observing response schedule-requirement has not been fulf i l l e d . Thus, l i t t l e information about the current food schedule would be conveyed in this case. In the Dinsmoor et a l . (1972) experiment, a reinforced observing response changed the colour of both the observing and food keys for 30 9 seconds* The rates of responding on the observing key were compared when S+ and/or S- were available as reinforcement for an observing response* Four conditions were studied, each of which had a different outcome of observing: 1) both S+ and S- available; 2) only S+ avail-able; 3) only S- available; 4) neither S+ nor S- available. Dinsmoor et a l . reported that observing behaviour was maintained under the fi r s t two conditions, but not under the latter two. An observing response which had S- as its only consequence rapidly extinguished. The differential reinforcing properties of S+ and of S- have been further examined by Jenkins and Boakes (1973). They reported that pigeons prefer a source that displays stimuli uncorrelated with reinforcement and extinction to a source presenting S- alone. Pigeons were givena choice between approaching an informative source, or one that was uncorrelated with food. When the outcome of the Informative source was restricted to S-, the pigeons showed a preference for the uncorrelated source. Mulvaney et al. (1974) have shown that a source of information with S+ a s i t s only consequence i s prefered over a source with both S+ and S- available. Three keys were concurrently available, the outer two of which were observing keys. Responding on one of the observing keys changed the colour of a l l keys to that associated with S+ or with S-, whichever was currently in effect. Responding on the other obser-ving key produced only the colour associated with S+. The pigeons responded at a much higher rate to the source associated with S+ only. Thus, in several recent, well-designed experiments, a general suppressive effect upon observing response rates by the S- has been 10. found. These findings lend considerable support to the conditioned reinforcement explanation of observing behaviour. 3. Purpose of Present Research The suppressive ef foot upon observing behaviour by S- in the Dinsmoor et al. (1972). Jenkins and Boakes (1973) and Mulvaney et al. (1974) experiments does not entirely preclude the possibility that S-may possess reinforcing Informative properties- as suggested ^Hendry (1969). If. as Sohaub (1969) has proposed, Sf and S- provide informa-tion of equivalent reinforcing value, i t would be the differential conditioned reinforcing value of the two stimuli, when summed with the reinforcing value of the Information they provide, that would determine the net reinforcing value of S+ and S-. For example, eons sider Condition 2 of the Dinsmoor et al. (1972) :«xperimentv/in3 which S- was the only consequence of the observing response. It may be that the aversive properties associated with this outcome are sufficient to overwhelm any reinforcing value arisingfrom the Informative- content of S-, with the result that observing behaviour is not maintained. There is considerable evidence which shows the punishing properties of a stimulus during which reinforcement is not available (cf. Herrnstein, 1955rlsrln'''-«nd--Hels.-'1966). On the other hand, S- may simply possess no reinforcing:Informa-tive properties, and this, irrespective of any punishing: aspect of S-, might be responsible for the failures to maintain observing behaviour with S- only. In the present experiment,-an effort is made to attenuate the punishing effect of S- to determine i f observing behaviour with S-11. as its sole consequence will then bo maintain As in the study by Dinsmoor et al. (1972). the design includes an observing-key on which S+ and/orS- are available according to a rand a food key oh which food reinforcement Is available on a random»interval schedule. In addition, a second food key is included^ on which reinforce-ment is also available on a random-interval schedule. Thus when an obser-ving response produces S-, indicating extinction on the fi r s t food key, the animal may continue to respond to the other food key, on which food-reinforcement availability is not affected by observing behaviour. Lei-tenberg, (1965), and later* Mulvaney et a l * (197*0 have noted; that i f S-is to retain Itsidehtity as a negative dlscrlmlnitive stimulus, the primary reinforcer must be withheld when i t i s present. However, the independence of the second food key from the observing and fi r s t food keys created by a changeover delay (Catania, 1966) insures that the identity of the S- will remain intact. Reinforcement on the second food key is available during either S+ or S- on the fi r s t key. Thus, the S-, by informing the animal that extinction is in effect on the fi r s t food key, retains the functionalattribute ofthe negative dlscrlmlnitive stimulus, but does not preclude the possibility of rein-forcement on the second food key* This procedure should attenuate the punishment associated with an observing response which produces S- only. Under this arrangement* the reinforcing value of the information provided by S- should become manifest, and, according to the information hypoth-esis, by itself maintain observing behaviour. This design does not alter the prediction based on the conditioned reinforcement hypothesis, viz.. that only the S+, whether alone or in conjunction with the S-, will effectively maintain the observing response. 12 METHOD 1. Subjects Three adult White King pigeons with varied previous experimental experience were maintained at approximately 80$ of their free-feeding weights throughout the experiment by grain obtained during experimental sessions. The subjects had free access to water and grit in the home cage. 2. Apparatus A three-key pigeon chamber (Lehigh-Valley Electronics, Model 1519) served as the experimental space. A horizontal array of translucent response keys on one wall was back-illuminated by coloured light. The left key was always illuminated by yellow light, while the center and right keys were illuminated at different times by yellow, red, or green light. Directly below the center key was a grain-feeder which allowed 5.0-sec access to mixed grain during reinforcement periods* A lamp illuminated the grain when the feeder operated. Another lamp provided general illumination of the chamber at a l l times, except during reinforce-ment periods. Solid-state and electromechanical circuits automatically scheduled experimental events and recorded data. White noise and the chamber air blower were used throughout the experiment to mask extrane-ous sounds. 3. Procedure Experimental sessions began at about the same time each day and lasted 60 min. With occasional exceptions, sessions occurred seven days 13. per week. During these sessions, the pigeons received preliminary-training, followed by the six experimental conditions outlined in Table 1 (pigeon #17 received only the fi r s t three treatments, due to an injury). Preliminary training. The pigeons were exposed to a variety of procedures that culminated in the arrangements described in Condition 1 below. Bird #17 underwent a total of 116 preliminary training ses-sions, while birds #19 and #20 received 57 and 40 preliminary sessions, respectively. ConditionH. observing responses produce S+ and S-„ A random-interval (RI) 120-sec (t=3; prr.025) schedule was in effect on the left key, under which food reinforcement for key pecking became available every 3 sec with a probability of .025. This key was always illumi-nated with yellow light. The center key was normally yellow, with food reinforcement for pecking available on a Mix RI 60-sec (t=3; p=.050) Ext schedule. Under the mixed schedule, there was an unsignalled alternation between periods of reinforcement availability (RI 60-sec) and extinction (Ext). The mean period length was 60 sec; individual periods were 30, 60 and 90 sec in duration. The right key was the observing key. Pecking this key, which was normally yellow, changed the colour of both this and the center key for a period of 23 sec, according to a RI 36-sec (t=3; pe.083) schedule. The tw6 keys became green (S+) during periods of reinforcement availability on the center key, or red (S-) during periods of extinction on the center key. By signalling reinforcement and extinction periods on the center key. TABLE 1 Summary of Procedure Schedule Number of Sessions Condition Left Key Center Key Right Key Bird #17 #19 #20 1. RI 120-sec, food Mix or Mult RI 60-sec Ext, food RI 36-sec, s+.s- 16 14 19 2, RI 120-sec, food Mix or Semi-Mult RI 60-seo Ext, food RI 36-see, S- 7 10 10 3. RI 120-sec, food Mix or Mult RI 60-sec Ext, food RI 36-sec, s+, S- 7 10 10 4. RI 120-sec, food Mix or Semi-Mult RI 60-seo Ext, food RI 36-sec, S- — 7 7 5. RI 120-sec, food Mix or Semi-Mult RI 60-sec Ext, food RI 36-sec, S+ — 7 14 6. RI 120-sec, food Mix or Mult RI 60-sec Ext, food RI 36-sec, s+,s- — 10 13 15. the observing response changed the mixed schedule to a multiple schedule (Mult RI 60-sec Ext), If the schedule of reinforcement changed during the 23-see observing period, the key colour changed accordingly. Condition 2: observing responses produce S- only* Food rein-forcement was again available on the lef t key according to a RI 120-sec schedule, and on the center key according to a Mix RI 60-sec Ext sched-ule. Pecking the observing key changed the colour of both the right and center keys to red (S-) according to a RI 36-sec schedule when extinction was in effect on the center key. Green (S+) was omitted as an outcome of observing• During periods of reinforcement availa-b i l i t y on the center key, when in Condition 1 an observing response would have changed the colour of the two keys to green, the keys simply remained yellow. By signalling only the extinction component on the center key, an observing response changed the mixed schedule to a semi-multiple schedule (Semi-Mult RI 60-sec Ext). The duration of the obser-ving response-produced S- was usually 23 sec. However, i f the schedule of reinforcement changed during this period, the two keys reverted to yellow* and the S- was shorter than 23 sec. Condition 3t observing responses produce Sf and S-y This con-dition was identical to Condition 1. Condition 4; observing responses produce S- only. This condi-tion was identical to Condition 2. Condition 5: observing responses produce Sf only. Reinforcement availability on the l e f t and center keys was identical to Condition 2. The observing requirement on the right key was again a RI 36-sec sched-ule. However, observing responses changed the right and center keys to 16. green (S+) during periods of reinforcement availability on the center key (Semi-Mult RI 60-sec Ext). Red (S-) was omitted as an outcome of observing. When extinction was programmed on the center key. peeking the observing key had no effect; both keys simply remained yellow. If the schedule of reinforcement changed during the 23-sec observing period, both keys reverted to yellow. Condition 6; observing responses produce S+ and S-. This con-dition was identicalto Condition 1. A changeover delay was used throughout- the six conditions, so that scheduled reinforcement or S+ or S. could not occur during the f i r s t 1.0 sec following a shift from one key to another. RESULTS Table 2shows the number of responses on the three keys, the num-ber of food reinforcements obtained on the le f t and center, keys, the number of stimuli (S+, S-) produced by the observing response, and the number of chahgebvers, summed across the last five sessions for each of the six conditions. Left food key. No systematic change in responding or in the num-ber of reinforcements obtained occurred across conditions. Center food key. Responses on the center key in the; presence of the observing stimuli (multiple schedule) are represented by G+ (S+) and R- (S-). The difference between G+ andR- responding in Conditions 1, 3 and 6 shows a near-perfect discrimination. High rates were maintained during G+, with l i t t l e responding during R-. For Birds #19 and #20, there was no systematic change in the rate of responding during G+ over TABLE 2 Data summed across last 5 days of each condition for each bird Center Key Left Key Responses Pood Right Key Condition Responses Food Reinforcera Y+ Y- G+ R- Reinforcers Responses S+ and/or S- Changeovers Bird #17 1. 4575 140 1034 1854 1026 2 140 619 129 2724 2. 6169 157 2271 2524 0 2 150 82 6 3397 3. 4726 133 1832 2228 1082 7 139 343 106 3363 Bird #19 i . 5095 147 1211 1427 2562 2 145 924 204 2638 2. 4395 134 2249 2463 0 7 126 #6 • 9 2246 3. 3547 141 1753 1526 1778 6 144 671 128 2360 4. 6863 153 3804 4340 0 0 164 110 10 3380 5. 4326 148 1410 1496 I896 0 140 919 124 3092 6. 7195 149 1967 1993 4643 4 148 1734 348 4150 Bird #20 1. 7846 150 1735 1494 4972 18 150 3178 331 3636 2. 10555 148 5163 5997 0 1 142 111 9 4506 3. 8651 148 2777 1970 4421 3 146 1755 241 4407 4. 9652 147 4857 5156 0 3 141 17 ,2 4104 5. 14©96i» 149 2749 4193 2054 0 145 1261 107 4519 6. 12086 150 3533 4974 1459 5 147 710 148 5028 18. Conditions 1, 3» 5 and 6» when that stiwultis was available as an out-come of observing. The presence or absence of R- in those conditions had no systematic effect upon responding during G+. Y+ and Y- indicate unsignalled periods of reinforcement availa-b i l i t y and extinction, respectively (mixed schedule). There was no systematic difference in responding during Y+ and Y-. Combined respond-ing during Y+ and Y- was appreciably higher in Conditions 2 and 4. Dor-, ing those conditions, G+ was not available as an outcome of observing, and since the pigeons were not responding for R-, there was akgreater opportunity for responding to Y-f and Y-. There was no systematic change in the number of food reinforcements obtained across conditions. Right (observing) key. The total number of observing responses during the last five sessions of each condition is shown in Table 2. Figure 3 shows responding on the observing key during each of these five sessions. A relatively high rate of responding occurred when both S+ and S- were available as outcomes of observing (Conditions 1, 3 and 6), and when only S+ was available (Condition 5)• For both Birds #19 and #20, responding for S+ only in Condition 5 was in the same range as responding for both S+ and S- in Conditions 1, 3 and 6, but not greater. When S+ was eliminated as an outcome of observing (Conditions 2 and 4), the observing rate underwent an immediate decline,; and re-mained at a very low residual level for the remaining sessions in those conditions. Observing behaviour was thus maintained by the production of both S+ and S-, and by S+ alone, but not by S- alone* Changeovers. There was no systematic change in the number of changeovers across conditions. CONDITIONS DC X O DC 800 700 600 500 400 300 200 100 0 1 2 3 4 5 6 17 -e-19 H > A 20 - A -w f V \ 1 A a i i i i i 11111 i i i i i LAST 5 SESSIONS Figure 3 s Last 5 sessions of responding, on the obser-ving key during the six treatments. Conditions are as follows: Observing responses (RO* s)>produce 1) both S+ and S-, 2) S- only, 3) both S+ and S-, k) S- only, 5) S+ only, 6) both S+ and S-. DISCUSSION In the present experiment i t was found that (1) pigeons observed for both S+ and for S+ only, but not for S- only, and (2) the ob-serving rate for S+ only was roughly equivalent to that for both S+ and S-. The failure to maintain observing behaviour when only S- was available as a consequence of observing can be reviewed within the frameworks of three hypotheses examined earlier. Information hypothesis. S-, indicating that extinction is in effect on the f i r s t food key, provides an equivalent asaount of infor-mation as S+, and should therefore be equally as effective in main-taining observing behaviour, i f , a3 Hendry (1969) proposed, i t i s the information provided by the observing stimuli that reinforces the ob-serving response. The results of the present experiment do not support such a notion, Two-faetor hypothesis. If, as Schaub (1969) has proposed, the reinforcing value of information provided by S+ and S- is equal, and sums with the respective conditioned reinforcing values of the two stimuli to determine the net reinforcing value of each, countering the negative conditioned reinforcing effect of S- should thus sustain re-sponding for S- only. In the present experiment, the addition of the second food key was intended to counter the negative effect of S- by providing an alternative source of reinforcement that was available irrespective of the schedule component in effect on the f i r s t food key. However, this procedure did not result in the maintenance of observing behaviour for S- only. It is possible that the suppressive effect of 21. S- on observing i s of sufficient magnitude to overwhelm any such ef-fort, in which case some reinforcing value of information could s t i l l be attributed to S-. While the present results cannot refute Schaub*a proposal, there is no evidence to favour a two-factor hypothesis over the more parsimonious conditioned reinforcement explanation. Conditioned reinforcement hypothesis. Wyckoff»s (1952; 1959; 1969) explanation of observing ascribes reinforcing properties to those stimuli associated with reinforcement, but not to those associated with nonreinforeement. Thus, S+ should maintain observing behaviour while S- should not. The present results, as well as the results of other recent experiments (Dinsmoor et a l . , 1972; Jenkins and Boakes, 1973; Mulvaney et a l . , 197*0 support this hypothesis. Further, the only exist-ing evidence contrary to this explanation (Schaub, 19^ 9; Lieberman, 1972) has been widely criticized on procedural grounds. Some comment should also be made on observing when S+ wa3 the only outcome of observing. There is substantial evidence showing that pigeons normally respond more when an observing response produces only S+ than when i t produces both S+ and S- (cf. Dinsmoor et al».1972; Mul-vaney et a l . . 197*0. Such an effect has been assumed to -be-due-to the inclusion of S- as an outcome of observing, which results in intermittent punishment of observing. In the present experiment, responding on the observing key for S+ only was in the same range as responding for both S+ and S-. It may be that the presence of the second food key, by atten-uating the suppressive effect of S- upon observing, accounted for the similarity in response rates between the two conditions. The present findings are consistent not only with Wyckoff»s (1952; 1959; 1969) conditioned reinforcement hypothesis of observing, but also 22 with a recant, and perhaps more parsimonious account of observing form-ulated by Browne and Dinsmoor (1974) after the present research had f begun. These authors note that in most observing experiments, the subject i s forced to spend equal amounts of time in the presence of S+ and S-, and that there is no means by which the animal can selec-t ively observe S+ and S_„ The only alternatives available are either to respond on the mixed schedule, or to respond on the observing key to change the mixed schedule to a multiple schedule, under which S+ and S-appear for equal periods. The fact that subjects respond on the observing key when S+ and S- are equally l ikely outcomes creates an explanatory problem, since observing behaviour cannot increase the frequency of food reinforcement, and since the mixed stimulus is associated with the same number of reinforcements per hour as S+ and S-. The result of this has been the propagation of the three theories already examined, Browne and Dinsmoor argue that the explanatory problem created by observing behav-iour may be only apparent, simply because of the way observing behaviour i s typically studied, Browne and Dinsmoor reported a procedure which allowed pigeons control over the duration of S+ and S-. An observing response was made by standing in one half of the chamber, S+ or S- could be sustained by remaining in that half of the chamber. Standing in the other half of the chamber produced the mixed stimulus. These authors found that pigeons spent far more time in the observing half of the chamber during S+ than during Such results can be explained by the matching law (Herrnstein, 1970), and, as Browne and Dinsmoor note, no appeal to infor-mation content, conditioned reinforcing value or similar hypotheses is necessary. The matching law states that organisms choose between 23. alternative reinforcement conditions in a manner such that the time spent in a condition matches the relative reiaforoement rate in that condition. Thus, organisms choose S+ over a mixed stimulus because the rate of reinforcement i s higher during the former than during the latter. Similarly, more time i s spent in the presence of the mixed stimulus than in the presence of S-, since the rate of reinforcement is higher in the former. The present finding that observing was maintained by S+ only but not by S- only can be explained by the matching law, since the former but not the latter stimulus was associated with a higher rate of reinforcement than the mixed stimulus. • The finding that observing was maintained when S+ and S- were equally l ikely outcomes can also be explained by Browne and Dinsmoor's proposal that the only reason the birds spent an equal amount of time in S+ and S- was because they were forced to do so. Had the birds been given a choice, they would have spent very l i t t l e time in S- when i t was produced by the observing response. Thus, i t appears that Browne and Dinsmoca^ s analysis of observing behaviour i s adequate for explaining the present results, and i t enjoys the further advantage of being the more parsimonious of existing accounts of observing., '^.;_-::^ .:. /'''-.•^ •^  24. REFERENCES Azrln, N.H. and Hole, W.C. Punishment. In W.K. Honig (Ed.). Operant  behavior: areas of research and application. New York: Apple-ton-C'entury-Qrofts, 1956. Pp. 380-447. Berlyne. D.E. Uncertainty and conflict: a point of contact between information theory and behavior concepts. Psychological Review. 1957. 64, 329-333. *~ Berlyne, D.E. Conflict, arousal and curiosity. New York: McGraw-Hill, . 1 96o. - — • Browne, M.P. and Dinsmoor, J.A. Wyckoff's observing response: pigeons learn to observe stimuli for free food but not stimuli for extinc-tion. Learning and Motivation, 1974, 5, 165*173. Catania, C.A. Concurrent operants. In W.K. Honig (Ed.), Operant behavior: areas of research and application. New York: Apple-lon-eeniury-crorts, 1900. Pp7"S.3-270. Dinsmoor, J.A., Browne, M.P. and Lawrence, C.E. A test of the negative discriminitive stimulus as a reinforcer of observing. Journal of the Experimental Analysis of Behavior, 1972, 18, 79-85. Dinsmoor, J.A., Flint, G.A., Smith, R.F. and Vlemeister, N.F. Differ-ential reinforcing effects of stimuli associated with the presence or absence of a schedule of punishment. In D.P. Hendry (Ed.), Conditioned reinforcement. Homewood, Illinois: Dorsey, 1969. Pp. 357-3&V Hendry, D.P. Reinforcing value of information: fixed-ratio schedules. In D.P. Hendry (Ed.), Conditioned reinforcement. Homewood, Illinois: Dorsey, 1969. Pp. 300-341. Herrnstein, R.J. Behavioral consequences of the removal of a discrimi-nitive stimulus associated with variable-interval reinforcement. Unpublished doctoral dissertation, Harvard University, 1955« Herrnstein, R.J. On the law of effect. Journal of the Experimental  Analysis of Behavior, 1970. 13. 24£2587~" ~ Jenkins, H.M. and Boakes, R.A. Observing stimulus sources that signal food or no food. Journal of the Experimental Analysis of Behavior, 1973. 20, 197-207.- ' Kendall, S.B. and Gibson, D.A. The effects of discriminative stimulus removal on observing behavior. Psychological Record, 1965. Iff. 545-551. 25. Leltenberg, H. Is time-out from positive reinforcement an aversive event? A review of the experimental evidence. Psychological  Bulletin. 1965, 64, 428-441. r;> Lieberman, D.A, Secondary reinforcement and information as determi-nants of observing behavior in monkeys. Learning and Motivation. 1972, 2 , 341-358 Mulvaney, D.B., Dinsmoor, J.A., Jwaideh, A.R., and Hughes, L.H. Punish-ment of observing by the negative discriminative stimulus. Journal of the Experimental Analysis of Behavior. 1974. 21, 37-44. Schaub, R.E. Response-cue contingency and cue effectiveness. In D.P. Hendry (Ed.), Conditioned reinforcement. Homewood. Illinois; Dorsey* 1969. Pp. 342-356. Wilton, R.N. and Clements, R.O. The role of information in the emission of observing responses; A test of two hypotheses. Journal of the Experimental Analysis of Behavior* 1971a. l6, 161-166. Wilton, R.N. and Clements, R.O. Observing responses and informative stimuli. Journal of the Experimental Analysis, of Behavior. 1971b, 1£, 199-204. Wyckoff, L.B. Jr. The role of observing responses in discrimination learning, part 1. Psychological Review. 1952. 59, 431-442. Wyckoff, L.B. Jr. Toward a quantitative theory of secondary, reinforce-ment, ftjyefoolojgj^ Wyckoff, L.B. Jr. The role of observing responses In discrimination learning. In D.P. Hendry (Ed.). Conditioned reinforcement. Homewood, Illinois: Dorsey, 1969. Pp. 237-260. 


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items