@prefix vivo: . @prefix edm: . @prefix ns0: . @prefix dcterms: . @prefix skos: . vivo:departmentOrSchool "Arts, Faculty of"@en, "Psychology, Department of"@en ; edm:dataProvider "DSpace"@en ; ns0:degreeCampus "UBCV"@en ; dcterms:creator "Spetch, Marcia Louise"@en ; dcterms:issued "2010-03-09T18:26:27Z"@en, "1979"@en ; vivo:relatedDegree "Master of Arts - MA"@en ; ns0:degreeGrantor "University of British Columbia"@en ; dcterms:description "In the present experiments stimulus control of pigeons' keypecking by the duration of food access was compared with control by the duration of light. These experiments were designed to extend previous research on three topics within the stimulus control literature: 1) stimulus control by food, 2) comparisons between the effectiveness of different stimuli in controlling behavior, and 3) control by stimulus duration. In Experiment 1, a go/no-go procedure was used to compare control of pigeons' keypecking by food-access duration with control by light duration. Pecks to an illuminated key were reinforced with grain following 10-sec presentations of food access or houselight, but not after 5-sec presentations of either stimulus. Each of the five subjects discriminated food-access duration faster and to a consistently greater degree than light duration. In four between-subject systematic replications, pigeons discriminated food-access duration better than the duration of a localized light, the feeder light, and a keylight, and with either water or food as reinforcement. Experiment 2 demonstrated the generality of these results across discrimination procedures. In this experiment, control by food-access duration was compared with control by light duration under a conditional right-left choice procedure (two pigeons), and a delayed symbolic matching-to-sample procedure (six pigeons). Under both, accuracy was greater on food-access trials than on light trials. In Experiment 3, two pigeons received generalization tests with durations of food access and light that were intermediate to the training values. The results showed that responding was controlled by the duration dimension for both stimuli. The excellent control by food access in these experiments is consistent with previous evidence that food is an effective and memorable stimulus for animals, and the difference in control by the duration of food and light provides empirical support for the common assumption that stimuli differ in their effectiveness of control. Several advantages of using a duration paradigm to compare control by different stimuli are discussed. This research, which demonstrates that the stimulus can play an important role in the accuracy of duration discriminations, also has implications for the study of control by stimulus duration, and for theories of timing processes in animals."@en ; edm:aggregatedCHO "https://circle.library.ubc.ca/rest/handle/2429/21674?expand=metadata"@en ; skos:note "PIGEONS DISCRIMINATE DURATIONS OF FOOD ACCESS BETTER THAN DURATIONS OF LIGHT by MARCIA LOUISE SPETCH B.A., The University of British Columbia, 1977 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF ARTS in THE FACULTY OF GRADUATE STUDIES (Department of Psychology) We accept this thesis as conforming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA September 1979 (c) Marcia Louise Spetch, 1979 I n p r e s e n t i n g t h i s t h e s i s i n p a r t i a l f u l f i l m e n t o f t h e r e q u i r e m e n t s f o r an a d v a n c e d d e g r e e a t t h e U n i v e r s i t y o f B r i t i s h C o l u m b i a , I a g r e e t h a t t h e L i b r a r y s h a l l make i t f r e e l y a v a i l a b l e f o r r e f e r e n c e a n d s t u d y . I f u r t h e r a g r e e t h a t p e r m i s s i o n f o r e x t e n s i v e c o p y i n g o f t h i s t h e s i s f o r s c h o l a r l y p u r p o s e s may be g r a n t e d by t h e Head o f my D e p a r t m e n t o r by h i s r e p r e s e n t a t i v e s . I t i s u n d e r s t o o d t h a t c o p y i n g o r p u b l i c a t i o n o f t h i s t h e s i s f o r f i n a n c i a l g a i n s h a l l n o t be a l l o w e d w i t h o u t my w r i t t e n p e r m i s s i o n . D e p a r t m e n t o f _ The U n i v e r s i t y o f B r i t i s h C o l u m b i a 2075 W e s b r o o k P l a c e V a n c o u v e r , C a n a d a V6T 1W5 n.tP SEPT An DE-6 BP 75-51 1 E ABSTRACT In the present experiments stimulus control of pigeons' keypecking by the duration of food access was compared with control by the duration of l i g h t . These experiments were designed to extend previous research on three topics within the stimulus control literature: 1) stimulus control by food, 2) comparisons between the effectiveness of different stimuli in controlling behavior, and 3) control by stimulus,, duration. . In Experiment 1, a go/no-go procedure was used to compare control of pigeons' keypecking by food-access duration with control by light duration* Pecks to an illuminated key were reinforced with grain following 10-sec presentations of food access or houselight, but not after 5-sec presentations of either stimulus,*. Each of the five subjects discriminated food-access duration faster and to a consistently greater degree than light duration*. In four between-subject systematic replications, pigeons discriminated food-access duration better than the duration of a localized light, the feeder light, and a keylight, and with either water or food as reinforcement. Experiment 2 demonstrated the generality of these results across discrimination procedures*, In this experiment, control by food-access duration was compared with control by light duration under a conditional right-left choice procedure (two pigeons), and a delayed symbolic matching-to-sample procedure (six pigeons). Under both, accuracy was greater on food-access t r i a l s than on light t r i a l s . In Experiment 3, two pigeons received generalization tests with durations of food access and light that were intermediate to the training values. The results showed that responding was controlled by the duration dimension for both stimuli. The excellent control by food access in these experiments i s consistent with previous evidence that food i s an effective and memorable stimulus for animals, and the difference in control by the duration of food and light provides empirical support for the common assumption that stimuli differ in their effectiveness of control,. Several advantages of using a duration paradigm to compare control by different stimuli are discussed* This research, which demonstrates that the stimulus can play an important role in the accuracy of duration discriminations* also has implications for the study of control by stimulus duration, and for theories of timing processes in animals. iv TABLE OF CONTENTS Abstr act .......................... ......... i i List of Tables ................... v List of Figures ... i . . . . . . . . . . . . . . . . . . . i v Acknowledgements ...... ................ . ... . .. . vi Introduction ............................. ..,,,,..» . 1 Experiment 1 .,i. .................. ............. ..... 14 Experiment 2 i . . . ... i ... .... .I.,.-..-.,. 31 Experiment 3 , i * , . . i ..i,i,;.ii ........ . . . . ,.i ... i * . .. .43 General Discussion ........................ ..... ... ... ,48 Reference Note ............ ..i 1... . i . - , . , . . i i 56 References ........................ ... «.i i i 57 LIST OF TABLES v Table 1, Subjects and Conditions of the Main Experiment and Four Replications . L . . . . . . . . . . . . . 22 Table 2. Mean Number of Keypecks During S+ Trials ........... 29 v i LIST OF FIGURES F i g u r e 1. Schematic diagram of the b a s i c procedure used i n Experiment 1, . • • • • . • • • . . 2 0 F i g u r e 2. D i s c r i m i n a t i o n r a t i o s (S-/S- + S+) on food-access and h o u s e l i g h t t r i a l s as a f u n c t i o n of t r a i n i n g f o r the f i v e pigeons i n Experiment 1. ..................................... 24 F i g u r e 3i. . D i s c r i m i n a t i o n r a t i o s on food-access and l i g h t t r i a l s as a f u n c t i o n of t r a i n i n g f o r pigeons i n the f o u r r e p l i c a t i o n s of Experiment 1. . . . . . . . . . . . . . . . . . ..... ............. 27 F i g u r e 4. Choice accuracy (percent c o r r e c t ) on food-access and h o u s e l i g h t t r i a l s as a f u n c t i o n of t r a i n i n g f o r the two pigeons i n Experiment 2A. . . . .. ................ . ......... ........ 36 F i g u r e Si. Matching accuracy (percent c o r r e c t ) on food- access and l i g h t t r i a l s as a f u n c t i o n of t r a i n i n g f o r the s i x pigeons i n Experiment 2B. ...... ............. .... .. .. .......... ..41 F i g u r e 6i P r o p o r t i o n of l e f t - k e y responses as a f u n c t i o n of stimulus d u r a t i o n f o r the two pigeons i n Experiment 3. . . . . . . . . . . .............. 46 ACKNOWLEDGEMENTS I would like to extend special thanks to my supervisor, D. M. Wilkie, for his support and valuable assistance with each stage of this research, I also wish to thank J. .P. „J. Pinel, R. W. Skelton, and L, J. Terlecki for their helpful comments and editorial suggestions; fi>.Summers, B,i Kirkbride, D> King, L. Leader, and M,. Gordon for their help in testing the subjects; and Ci Spetch for typing assistance,. 1 INTRODUCTION I t i s often assumed that some s t i m u l i are more e f f e c t i v e than others i n c o n t r o l l i n g a given organism's operant behavior (e.g». Mackintosh, 1977). More s p e c i f i c a l l y , some authors (e-g., Staddon, 1974) have suggested that food i s a p a r t i c u l a r l y s a l i e n t and memorable stimulus for hungry animals. The present research was designed to provide some empirical evidence for this assumption. In the present experiments stimulus control of pigeons' keypecking by the duration of food access was investigated and compared with control by the duration of l i g h t . In the next three sections of the Introduction, stimulus control and antecedent simulus control are defined, and the experimental procedures commonly used to establish and study them, are described. A review of the relevent l i t e r a t u r e on antecedent control by stimulus duration comprises the f i f t h sectionw Although control by the duration of a variety of stim u l i has been demonstrated frequently, no e x p l i c i t comparisons of control by the duration of d i f f e r e n t stimuli have been reported* In the sixth section, empirical evidence for the common assumption that s t i m u l i d i f f e r in t h e i r a b i l i t y to acquire control of behavior i s examined and found to be lacking. Several attempts to compare d i f f e r e n t stimuli have produced ambiguous r e s u l t s because a diff e r e n t dimension was used for each stimulus and the values selected from these dimensions not equated for d i s c r i m i n a b i l i t y . In order to compare s t i m u l i , one must either equate the values of the different dimensions by psychophysical scaling techniques, or use a dimension common to both s t i m u l i as the basis for the comparison. A discussion of 2 the p o s s i b i l i t y of using duration as a common dimension with which to compare stimulus control by various s t i m u l i comprises the seventh section, and in the eigth section, evidence consistent with the suggestion that food i s an e f f e c t i v e c o n t r o l l i n g stimulus i s described. The l a s t section of the Introduction i s a summary of the rationale and purpose of the present research. Stimulus Control of Operant Behavior In operant conditioning, reinforcement i s made contingent upon a class of responses.. Inevitably, certain s t i m u l i , such as l i g h t or sound, are present when these responses occur. Sometimes these stimuli acquire control of these responses; i . e . , changes i n the stimuli (e.g., t h e i r presentation) r e s u l t i n changes in some response c h a r a c t e r i s t i c (e.g., rate of occurrence). When changes i n responding covary i n a regular fashion with changes i n a stimulus along a given dimension, control by the stimulus dimension i s said to exist. Both the occurrence and nonoccurrence of such stimulus control apparently can be adaptive to an organism.. For example, i t i s easy to imagine how control of pigeons' consumatory responses by the v i s u a l aspects of food and water might be adaptive;. A pigeon that approaches and pecks at objects that resemble grain but not those that resemble stones l i k e l y would survive longer than a pigeon that responds i n the same manner to both. Such differences i n behavior i n the presence of d i f f e r e n t s t i m u l i exemplify discrimination. In other s i t u a t i o n s , discrimination would not be adaptive. For example, pigeons 3 approach and peck grains of various shapes and colors as well as grains found i n new locations.. This generalization of behavior to s i m i l i a r but novel situations undoubtedly increases the pigeon's chances of obtaining an adequate supply of food. Establishing Stimulus Control Operant discrimination t r a i n i n g i s a procedure s p e c i f i c a l l y designed to establish stimulus control* By using t h i s procedure, one can experimentally examine the ef f e c t of variables on stimulus control. In discrimination t r a i n i n g , reinforcement i s made contingent upon responding i n the presence of a certa i n stimulus, often c a l l e d the discriminative stimulus (Skinner, 1938). For example, a pigeon's keypecking may be reinforced i n the presence of one stimulus (S +) but not reinforced in i t s absence or in the presence of another (S~) . The usual r e s u l t of such t r a i n i n g i s a higher rate of responding during S+ than during S~. Control by Antecedent Stimuli A c o n t r o l l i n g relationship between st i m u l i and responding also may develop under conditions i n which the S* and S -terminate and are not present when responding occurs. This i s referred to as antecedent stimulus control* Operant discrimination procedures designed to establish such control are often c a l l e d memory paradigms because i n order for t h i s type of control to occur the organism must remember in some fashion the S+ and/or S~. Thus, studies of antecedent stimulus control are relevent to the growing l i t e r a t u r e on animal memory processes 4 (Honig & James, 1971; Hulse, Fowler, & Honig, 1978; Medin, Roberts, & Davis, 1976), antecedent s t i m u l u s c o n t r o l i s e s t a b l i s h e d by one of two c o n d i t i o n a l d i s c r i m i n a t i o n procedures.. Under qo/no-qo procedures, the occurrence o f S + or S - determines whether responding to a s t i m u l u s t h a t f o l l o w s both S* and S~ w i l l or w i l l not be r e i n f o r c e d . For example, W i l k i e and Wilson (1977) r e i n f o r c e d pigeons' keypecks to a white pecking key when i t was preceded by a white l i n e (S +) on a green surround p r o j e c t e d on the key, but not when i t f o l l o w e d green alone (S~) on the key.. As a r e s u l t , the l i n e S + c o n t r o l l e d responding t o the white key. Under c h o i c e procedures, one of a number of antecedent s t i m u l i i s f o l l o w e d by two or more c h o i c e s t i m u l i , fieinforcement f o r responding to some of these i s c o n t i n g e n t upon the preceding antecedent s t i m u l u s . T h i s c h o i c e procedure i s e x e m p l i f i e d by the delayed matching-to-sample (e,.g., Blough, 1959), delayed oddity-from-sample (Lyderson, P e r k i n s , & C h a i r e z , 1977) , and delayed symbolic matching-to-sample (e.g., W i l k i e , 1978) procedures, as well as the delayed s p a t i a l (e.g., Hearst, 1962) and n o n s p a t i a l (e.g., W i l l i a m s , 1971) a l t e r n a t i o n procedures,. Antecedent C o n t r o l by. Stimulus D u r a t i o n S e v e r a l experiments have i n v e s t i g a t e d antecedent c o n t r o l by s t i m u l u s d u r a t i o n ; For example, Reynolds and Catania (1962), using a go/no-gb procedure, demonstrated c o n t r o l of pigeons' keypecking by the d u r a t i o n of an i n t e r v a l during which a pecking key was dark,. In t h e i r study, the pecking key was i l l u m i n a t e d 5 for 30 sec following a variable dark-key duration that ranged from 3 to 30 sec in 3-sec increments,. Keypecking was reinforced according to a 20-sec variable-interval (VI) schedule after the 30-sec dark-key duration (S+), but not following shorter durations of the dark key (S-),. As a result, the rate of responding to the lighted key increased in a regular fashion with increases in the duration of the preceding dark-key interval. Stubbs (1968) demonstrated control of pigeons* choice behavior by stimulus duration using a delayed symbolic matching-to-sample procedure. Trials began with illumination of the center key with orange light. A peck at this key changed the stimulus from orange to white. The white key stimulus then remained on for one of several durations, half of which were designated as long, the other half, as short,. As soon as the center white stimulus terminated two side keys were illuminated, one with red light and one with green* Pecks to the red key were reinforced after any of one class (short or long) of white-key durations- Pecks tc the green key were reinforced after any of the other class of white-key durations;* As a result, the duration of tie white key controlled pecks to the red and green keys. Pigeons' choice behavior also has been controlled by the duration of a preceding tone (Kinchla, 1970) and by the duration of their own keypecks (Ziriax & Silberberg, 1978); and rats' choice behavior has been controlled by the duration of blackouts (Church & Deluty, 1977) and white noise (Church, Getty, & Lerner, 1976), 6 Most of these experiments have focused primarily on the psychophysical aspects of temporal discrimination. For example, various psychophysical procedures have been used to investigate difference limens (Church, Getty, 6 Lerner, 1976) and points of bisection for temporal inte r v a l s (Church & Deluty, 1977; Stubbs, 1976a),. Elsmore (1972), Kinchla (1970), and Stubbs (1976b) have analysed duration discrimination i n terms of the psychophysical theory of signal detection (e.g., Green & Swets, 1966),. The e f f e c t s of procedural variables on duration discriminations have been studied infrequently^ Elsmore (1971) found that discrimination t r a i n i n g was necessary for the development of stimulus control of pigeons' keypecking by blackout duration* Pigeons f i r s t were given n o n d i f f e r e n t i a l training i n which key illuminations alternated with blackouts of a single duration, and pecks to the illuminated key were reinforced under a VI schedule. A subsequent generalization test i n which various durations of blackout were presented showed no control of keypecking by blackout duration*. 1 However, following discrimination t r a i n i n g in which keypecking was reinforced following one blackout duration but not following another, generalization tests showed that blackout duration had 1 This means of assessing stimulus control after n o n d i f f e r e n t i a l reinforcement i s subject to certain problems*. For example. Mackintosh (1977, p,. 491) has pointed out that such f a i l u r e s to observe stimulus control may not be due to the n o n d i f f e r e n t i a l t r a i n i n g , but rather to masking by other s t i m u l i . In Elsmore*s study, the v i s u a l stimulus on the pecking key may have acquired s u f f i c i e n t l y high control of pecking to mask any control acguired by the blackout duration. Also, Wilkie and Masson (1976) have shown that f a i l u r e s to demonstrate stimulus control might be the r e s u l t of the testing procedures,*, 7 acquired control of keypecking* The procedure used to establish stimulus control affects the accuracy of temporal discrimination. Perikel, Richell, and Maurissen (1974) compared pigeons' duration discriminations under go/no-go versus spatial choice procedures,; The choice procedure resulted in smaller duration-difference thresholds than the go/no-go procedure. Although a variety of stimuli have been used in duration experiments, there have been no explicit comparisons of control by the duration of different stimuli. The lack of experimental attention to this variable i s surprising for two reasons,. First, the results of duration discrimination studies have been used to draw inferences about timing processes in animals (e.g., Eoberts & Church, 1978), and yet the generality of animals' duration discriminations across stimuli has not been established* Second, i t i s often assumed that stimuli differ in terms of saliency, or their effectiveness in controlling behavior (e*g, , Staddon, 1974); i f this i s true then i t i s possible that the extent of control by duration may vary with the stimulus. The empirical basis for the assumption that stimuli differ in effectiveness of control i s discussed and evaluated in the next section. Comparisons Between Stimuli for Effectiveness of Control Many investigators assume that certain stimuli acquire control of responding more readily and to greater degree than other stimuli, For example, Mackintosh (1977, p. 483) stated: \"the amount of training required to establish control by a particular stimulus will surely vary from 8 stimulus to stimulus*..No one would deny that some stimuli appear to be more effective for some subjects than are others.\" Although there are some experimental data consistent with this assumption, surprisingly few studies have provided direct evidence for differences between the effectiveness of different stimuli in controlling behavior, Several attempts to demonstrate that stimuli differ in their a b i l i t y to control behavior have involved comparisons across stimulus dimensions. For example, Chow (1953) compared monkeys' acquisition of a contour (circle vs* square) discrimination and a pattern (diamonds vs. stripes) discrimination using a Kluver form board, and reported faster acquisition of the contour discrimination* Carter and Eckerman (1975) compared control by two colors (red vs. green) with control by two line t i l t s (180° vs. 90°) in a matching-to-sample task with pigeons, Watching was established faster, and terminal accuracy was higher with colors than with line t i l t s . Unfortunately, the results of both of these studies are ambiguous for two reasons,* F i r s t , i t i s not clear what dimensions of the stimuli controlled behavior. For example, a discrimination between red and green could reflect control by any or a l l of a number of dimensions, such as wavelength, intensity, or saturation. Second, i t i s not clear whether the difference between the two values selected from each dimension were equal* Such differences are very important because i t i s known that the extent of control aquired by a given dimension varies directly with the difference between the values used as discriminative stimuli. For example, two line t i l t s differing 9 by 90° readily control pigeons1 keypecking whereas two line t i l t s differing by only 5° might not. Therefore, comparisons of different stimuli across stimulus dimensions require psychophysical isometric analysis to equate the differences between values from each dimension. One method for doing this would be to use psychophysical procedures to find the just noticeable difference (JND) along each dimension for the given species;. Then, values from each dimension that differ by the same number of JNDs could be selected to compare stimuli* The effectiveness of stimuli can be compared with less d i f f i c u l t y by using a dimension common to each stimulus as the basis of the comparison. A few comparisons of this sort have been attempted, but these too have suffered from problems* For example, Jarvik (1953) used wavelength as the common dimension to compare stimuli with a simultaneous discrimination procedure. In this study, control of monkeys' choice behavior by the color of bread was compared to control by the color of squares of celluloid. Under one condition, red and green squares of bread (either dyed bread, cr bread covered with colored celluloid) served both as the discriminative stimuli and as the reinforcers* Bread of one color was unflavored; the other was quinine-adulterated. The monkeys were permitted to choose and consume one of these,. The color of bread rapidly controlled choice behavior; nonei of the monkeys chose the quinine-adulterated bread more than once in 25 t r i a l s . The same subjects were tested in a second condition in which the discriminative stimuli were red and green squares of celluloid. A square of cne color was located in front of a piece of 10 unflavored white bread; a square of the other color was located in front of quinine-adulterated white bread,. The color of the celluloid acquired l i t t l e control of choice behavior; the monkeys chose the quinine-adulterated bread on almost half of the 25 t r i a l s * Because the spatial and temporal contiguity between the stimulus, response, and reinforcer were not the same in the two conditions, i t is impossible to determine whether the difference in control by the color of bread vs. the color of celluloid was due to the stimuli themselves, or to differences in the nature of the task* Furthermore, the dimensions that controlled responding to the red and green squares of bread or celluloid were not established* It i s possible for example, that the monkeys' responses were controlled by intensity differences between red and green; no attempt was made to equate the intensity differences in the two conditions. A close examination of similar attempts to compare stimuli (e.g., Deluias & Emmerton, 1978; Dobrzecka, Szwejkowska, & Konorski, 1966; Harrison 6 Briggs, 1977) reveals that these studies also did not establish which dimension(s) of the stimuli controlled responding* Thus, comparisons between the abil i t y of different stimuli to control behavior are not as easy as one might f i r s t imagine,. First, a dimension common to each stimulus must be selected, or isometric scaling must be conducted i f different dimensions are used,. Second, evidence must be provided that the nominal dimension(s), actually control (s) responding. This latter requirement i s discussed further in the introduction to 11 Experiment 3* Selecting a Common Dimension for Comparing Stimuli Many dimensions are common only to a limited range of stimuli* For example, color i s a property of only visual stimuli. Moreover, even when there are dimensions common to stimuli perceived by different sensory modalities, they are frequently not suitable for comparisons of such stimuli because they are measured in different units for each modality and therefore would require rescaling in isometric units, Intensity, for example, i s a property of both light and sound; however one cannot equate a particular sound intensity difference with a particular light intensity difference without elaborate psychophysical studies. One dimension that i s common to a l l stimuli i s duration. The values of this dimension can be equated accurately across stimuli, and can be manipulated easily within an experimental setting. Thus , the duration dimension seems well suited to comparisons among stimuli. One could use duration as the common dimension to compare a variety of stimuli, for example, lights and tones* If control by certain durations of light i s superior to control by the same durations of tones, or vice versa, this would imply that certain other properties of the stimuli differentiate them, For example, lights (or tones) may possess properties that enhance attention to, or exploration of, such stimuli, and this might increase the likelihood that the different values on the duration dimension would be detected. Additional research would be required to specify the properties 12 which enhance the ability of a stimulus to control behavior* Stimulus Control by Biologically Important Events If stimuli differ in effectiveness of control, i t i s possible that there might be large differences between stimuli that differ substantially in their biological importance to the organism;. Stimuli that are important for survival, such as food or water, commonly are used to reinforce behavior; i t would not be.surprising i f such stimuli also were very effective as discriminative stimuli. Staddon (1974), for example, has suggested that reinforcing stimuli possess \"intrinsic salience\" and may be \"more memorable\" than \"neutral\" stimuli,. In general, the evidence from studies that have investigated stimulus control by food or water presentations supports the assumption that these stimuli readily acquire control of behavior and are very memorable. For example, there is considerable evidence that the occurrence or nonoccurrence of food is a memorable event for hungry animals* Numerous reward alternation studies have demonstrated that the presence or absence of food reward will control rats 1 running speeds (e.g., Capaldi, 1967) and bar press latencies (e.g., Wall, & Goodrich, 1964) on the next t r i a l , even when the t r i a l s are separated by up to 24 hr (e.g., Capaldi & Spivey, 1964)* Similarily, the presence or absence of grain has controlled pigeons' keypecking over long delays in both delayed symbolic matching-to-sample tasks (Maki, Moe, & Bierly, 1977; Wilkie, 1978) and a response-independent go/no-go discrimination task (Botjer & Hearst, 1979)* 13 Several studies have shown that qualitative properties of reward also control behavior readily,. For example, Pschirrer (1972) found that type of reward (milk vs. food pellets) controlled rats' running speeds in an alternation paradigm, as well as their choice between two runways. Cruse, V i t e l l e l l i , and Dertke (1966) used a multiple schedule procedure in which a standard food pellet initiated one schedule of reinforcement, and a sucrose pellet initiated a different schedule, and found good control of rats' bar press rates and choice responses by the type of pellet; Finally, Spetch and Wilkie (Note 1), using a reward alternation paradigm, demonstrated control of pigeons' keypecking rates and topographies by type of reward (food vs. water),; In view of this growing body of evidence that reinforcing stimuli are very effective in controlling operant behavior, i t seemed worthwhile to directly compare control by such stimuli with control by a more \"neutral\" stimulus. Therefore, in the present experiments the effectiveness of food access and lights as discriminative stimuli were compared; The duration dimension served as the basis for this comparison. Purpose of the Present Research Antecedent stimulus control by the duration of food-access had not been investigated; thus, one purpose.of this research was to demonstrate and describe such stimulus control, In the present experiments stimulus control of pigeons' keypecking by the duration of food access was compared to control by the duration of lights. The purpose of these 14 comparisons was two-fold. F i r s t , these experiments sought to provide an uneguivocal demonstration of differences between stimuli in their effectiveness in controlling operant behavior. Second, these ccmpariscns were designed to investigate the importance of the stimulus in control by stimulus duration. For example, in previous stimulus duration experiments traditional \"neutral\" stimuli such as lights or noise have been used; control by the duration of these stimuli may differ in an important way from control by the duration of a more \"biologically important\" stimulus. Such a finding would have implications for the generality of the previous work on duration discrimination. EXPERIMENT 1 In this experiment food-deprived pigeons were exposed to an elaboration of the go/no-go duration discrimination procedure used by Reynolds and Catania (1962). Pecks to a key illuminated with green light were reinforced with grain when the key illumination was preceded by a 10-sec (S*) stimulus duration, but not when i t was preceded by a 5-sec (S-) stimulus duration. A grain-filled feeder (food access) served as the stimulus on half of the t r i a l s ; the houselight served as the stimulus on the remaining t r i a l s * This procedure allowed within-session, within-subject comparisons of the acquisition of control by food-access duration and houselight duration* In subsequent replications, control by food-access duration was compared to control by the duration of a localized light, a keylight, and the feeder light. These replications were 15 included to determine whether control by food-access duration would d i f f e r in a consistent way from control by the duration of various forms of l i g h t . In two of these r e p l i c a t i o n s , keypecks were reinforced with water instead of food, to determine whether such differences i n control were s p e c i f i c to food-reinforced behaviors. Experimental Design In t h i s experiment (and the others reported below) s i n g l e -subject methodology (cf,; Hersen & Barlow, 1976; Sidman, 1966) was employed. Single-subject methodology d i f f e r s from other experimental approaches i n two important ways; F i r s t , the data of i n t e r e s t are those of i n d i v i d u a l subjects, not average values of several indi v i d u a l s taken together; Second, the r e l i a b i l i t y and generality of the data from the individ u a l s i s assessed by r e p l i c a t i o n , either with the same or di f f e r e n t subjects (within-or between-subject r e p l i c a t i o n s , respectively) and with the same or s l i g h t l y d i f f e r e n t procedures (direct or systematic r e p l i c a t i o n s , r e spectively). In the present experiments, dir e c t between-subject and systematic within- and between-subject r e p l i c a t i o n techniques were employed. Method Subjects The subjects were six S i l v e r King (Birds 1, 2, 6, 7, 8, and 9) and f i v e White. King (Birds 3, 4, 5, 10, and 11) pigeons. 16 B i r d 7 was e x p e r i m e n t a l l y naive; B i r d s 3 and 4 had r e c e i v e d only autoshaping t r a i n i n g p r i o r t o t h i s experiment. The remaining s u b j e c t s had v a r i e d experimental h i s t o r i e s * A l l _ s u b j e c t s were deprived of food u n t i l they were between 80 and 85% of t h e i r f r e e - f e e d i n g weights. They were maintained at these weights throughout the study by the mixed g r a i n obtained d u r i n g t h e . experimental s e s s i o n s , and p o s t - s e s s i o n feedings of maple peas,. Some s u b j e c t s a l s o were water d e p r i v e d (see Table 1); d a i l y water i n t a k e was r e s t r i c t e d t o the 15 to 20 ml obtained d u r i n g the experimental s e s s i o n s * A l l b i r d s were housed i n l a r g e i n d i v i d u a l cages i n which h e a l t h g r i t always was a v a i l a b l e . S u b j e c t s t h a t were not water deprived had u n l i m i t e d access t o water i n t h e i r home cages. Apparatus A standard BBS-Foringer Model PS-004 one-key pigeon chamber was used d u r i n g the main experiment and E e p l i c a t i o n s 1, 2, and 3. The c l e a r p l a s t i c pecking key, which r e q u i r e d a f o r c e o f about 0.2 N to operate, was l o c a t e d on the center of one w a l l . An I n d u s t r i a l E l e c t r o n i c s Engineers' S e r i e s 10 s t i m u l u s p r o j e c t o r c o n t a i n i n g 2.8 W lamps was mounted behind the key and i l l u m i n a t e d the key with a uniform f i e l d of green l i g h t . A BSE/LVE Model #114-10 s o l e n o i d - o p e r a t e d g r a i n feeder t h a t allowed access t o mixed g r a i n was l o c a t e d d i r e c t l y below the key,* Feeder o p e r a t i o n s were accompanied by i l l u m i n a t i o n of a 2,. 8 W lamp i n the hopper. A BSR/LVE Model #114-06 s o l e n o i d -operated water d i s p e n s e r was mounted on the w a l l t o the l e f t of the feeder* A p l a s t i c c o n t a i n e r , 3.5 cm i n diameter, was 17 mounted on the wall below the water dispenser* This served as a rec e p t i c l e into which a measured volume of water could be dispensed* A 4.8 W lamp was mounted on the wall 3 cm to the l e f t of the water r e c e p t i c l e . The houselight consisted of two 2,8 W lamps i n the upper front corner of the chamber, mounted behind a metal panel. When these were l i t , r e f l e c t e d l i g h t from the panel to the white c e i l i n g provided d i f f u s e . i l l u m i n a t i o n of the entire chamber* A BES/LVE Model #132-02 two-key pigeon chamber was used during Replication 4,. The two clear p l a s t i c keys, which required a force of about 0*2 N to operate, were located on one wall 18 cm apart. An Electronics Engineers' Series 10 projector, containing 2,8 W lamps, was mounted behind each key. These illuminated the l e f t key with a uniform f i e l d of white l i g h t , and the right key with a uniform f i e l d of red l i g h t . A BRS/LVE Model #114-10 solenoid-operated grain feeder was centered on the same wall between and below the two pecking keys. Feeder operations were accompanied by illumination of a 2,8 W lamp in the hopper. Houselights were never used i n t h i s r e p l i c a t i o n , Experimental conditions and data recording were controlled by D i g i b i t s o l i d - s t a t e and electromechanical c i r c u i t s . Procedure General Procedures In t h i s experiment and a l l experiments reported below, each suject received one experimental session per day*. Sessions were conducted 7 days per week at approximately the same time each 18 day,. A l l subjects were weighed intermittently* _ Preliminary Training The preliminary training for a given subject depended upon i t s prior experimental experience and the experimental condition under which the subject would be tested,. a t the end of training, each bird reliably: (a) ate from the raised illuminated hopper and/or drank from the illuminated water recepticle (Replications 2 and 3 only) , and (b) pecked with a high and stable rate at the illuminated response key* Discrimination Training A) Main Experiment Figure 1 illustrates the basic discrimination procedure of the main experiment. a discrete t r i a l go/no-go procedure was used* The four types of t r i a l s each began with either houselight or feeder operation (food access). These stimuli remained on for either 10 sec or 5 sec. a 1.5-sec delay, during which the chamber was dark, followed* The pecking key was then illuminated with green light for 10 sec. Pecks were recorded but were without consequence for the f i r s t 8 sec of key illumination,, a peck that occurred during the last 2 sec produced reinforcement (5 sec of grain access) i f the preceding stimulus* either houselight or food access, had been 10 sec in duration (S+) * Pecks following 5-sec durations (S~) of food access or houselight were never reinforced. Trials were separated by a 50-sec i n t e r t r i a l interval* The.four types of t r i a l s occurred an equal number of times (approximately 10), in mixed order within a session. 19 Figure 1. Schematic diagram of the basic procedure used in Experiment U 2 0 HOUSELIGHT F E E D E R P E C K I N G K E Y 1 ^ 5 sec ^ | ^ 10 sec ON 10 sec 1 l < > | G R E E N LIGHT P R E S E N T E D D E L A Y INTERVAL (1 .5sec) . 1/ l < -P E C K S , 2 sec RECORDED R E I N F O R C E M E N T (5 sec GRAIN ) A V A I L A B L E ON 10 sec T R I A L S INTERTRIAL INTERVAL (50 sec) 21 B) Replications Four systematic replications of the main experiment were conducted,. In these, certain features of the discrimination procedure (Figure 1) were modified. Replication 1 involved a single subject; Replications 2, 3, and 4 each involved two subjects* Replication J Water reinforcement (1 ml of water was dispensed and the lamp 3 cm to the l e f t of recepticle was illuminated for 5 sec) was substituted for the food reinforcement. Replication 2 a localized light (illumination of the lamp located next to the water recepticle) was used in place of the houselight* Replication 3 The feeder light (illumination of the lamp in the hopper without presentation of grain) was substituted for the houselight* and water was substituted for food as the reinforcement* Replication 4 a keylight (illumination of the l e f t key with white light) was substituted for the houselight, and the right response key was illuminated with red rather than green light. The main experiment and a l l replications were terminated when discriminated responding appeared to be stable and asymptotic. Table 1 summarized these various experimental conditions and presents the number of sessions for each subject* 22 Table J Subjects and Conditions of the Main Experiment and Four Replications Experiment Bird Stimuli Reinforcement Deprivation Sessions Main 1 Food Access Food Food 50 vs Houselight II 2 \" \" \" 7 0 II 3 II II \" 5 0 II 4 II n \" 7 0 it 5 II II \" 7 0 Replication 1 6 Food Access Water Food 70 vs and Houselight Water Replication 2 7 Food Access Food Food 80 vs Localized Light II 8 \" 1 1 \" 100 Replication 3 7 Food Access Water Food 70 vs and Feeder Light Water 70 Replication 4 10 Food Access Food Food 100 vs Keylight II 11 II II II 100 23 Data Analysis Control of responding by the duration* of the food and the light stimuli was assessed by calculating a separate discrimination ratio for food-access and light t r i a l s . Pecks after the S~ duration of a stimulus (food or light) were divided by the sum of pecks after the S+ and s- durations of that stimulus. Thus, indiscriminate responding after the two durations would produce ratios of ,5 and the ratios would decrease as control by the stimulus duration develops* These ratios were calculated for each subject for blocks of 10 consecutive sessions and plotted to permit visual inspection of the acquisition of control by the duration of food access and light. Results Main Experiment The discrimination ratios for blocks of 10 sessions are shown in Figure 2, Individual pigeons* data are presented in separate panels* Acquisition curves for food-access t r i a l s are shown by closed circles; open circles show the acquisition curves for houselight t r i a l s . Two features are apparent in the data of a l l birds* First* the asymptotic values of the discrimination ratios for both the food-access and light stimuli were below ,5; thus, the duration 1 Justification for assuming that duration was the controlling dimension i s provided by Experiment 3* 24 Figure 2. Discrimination ratios (S-/S-+S+) on food-access and houselight t r i a l s as a function of training for the five pigeons in Experiment 1. (Lower-valued ratios indicate better discrimination of the stimulus durations). 25 BLOCKS OF 10 SESSIONS 26 of both stimuli acquired some control of the responding of a l l five birds* The second and most striking aspect of these data i s the difference in control by food-access and houselight duration. Without exception, the discrimination ratios for food-access t r i a l s decreased more rapidly and remained lower (i.e,, discrimination was better) than those for houselight t r i a l s . Food-access duration clearly was more effective in controlling responding than was houselight duration* Replications Figure 3 presents the acquisition curves for subjects in the four replications. Each replication involved a different light stimulus* Replications 1 and 3 involved water reinforcement, whereas Replications 2 and 4 involved food reinforcement* In spite of these differences, a l l subjects showed the same pattern of results; control by food duration developed more rapidly and to a greater degree than did control by light duration* Table 2 presents each subjects 1 rates of keypecking during food-access and light S+ t r i a l s for the main experiment and the replications. Differences in rates of pecking on these S+ t r i a l s did not seem to play an important role in the differences in control by these two stimuli* &11 subjects discriminated food-access duration better than light duration in spite of the fact that some birds responded more on food S+ t r i a l s and others responded more on light S + t r i a l s . 27 Figure 3 . Discrimination ratios on food-access and light t r i a l s as a function of training for pigeons in the four replications of Experiment 1. 2 REPLICATION 1 REPLICATION 2 REPLICATION 3 REPLICATION 4 a., •z. O .00-1 1 , 1 , \\— • 2 4 6 8 < BLOCKS OF 10 SESSIONS 29 Table 2 Mean Number of Keypecks During S* T r i a l s Experiment B i r d L i g h t S + Food-access S+ Main R e p l i c a t i o n 1 R e p l i c a t i o n 2 II R e p l i c a t i o n 3 1 2 3 4 5 7 8 7 9 R e p l i c a t i o n 4 10 6.54 5,. 94 15. 30 12, 33 9.59 14, 80 8.42 5.25 10*36 16. 26 17. 02 8*35 13. 26 5.32 20.39 9,57 1 l i 63 24*86 11,25 9*41 14.03 18,. 56 16; 33 6,97 30 Discussion These results demonstrate that food-access duration readily controls pigeons' keypecking under a go/no-go discrimination procedure. Such control appears to be gualitatively similar to control by other other types of stimuli; i,e*, discrimination improved as a function of training* Apparently, however, control of keypecking by food-access duration does differ quantitatively from control by light duration. Three features of the results provide evidence for the r e l i a b i l i t y and the generality of these differences. F i r s t , the superior control by food-access duration was replicated in every subject studied* Second, these differences in control were not restricted to food-reinforced behaviors. Water-reinforced keypecks (Replications 1 and 3) also were controlled more readily by food-access duration than by light duration. Third, the comparisons between food access and the four light stimuli yielded comparable results even though these lights differed in such features as intensity, l o c a l i z a b i l i t y , size, and location in the chamber* In Replication 3 (food access vs* feeder light), a l l features of the two stimuli were identical, except the presence and absence of food,. Therefore, the large difference in control by the duration of these stimuli seems to .be due to some property (ies) of food,. 31 EXPEHIMENT 2 Although Experiment 1 demonstrated that pigeons' keypecking in a go/no-go task i s controlled more readily by the duration of access to food than by the duration of light, the generality of this finding across procedures had to be investigated before general conclusions could be reached about the relative effectiveness of these stimuli in controlling pigeons' keypecking;. In fact, there i s evidence which suggests that the relative control acquired by different stimuli can vary with different operant discrimination procedures,. For example, several experimenters (e.g., Dobrzecka, Szwejko^wska, & Konorski, 1966; Lawicka, 1966) have reported that the position, but not the quality, of two auditory stimuli (buzzer vs, metronome) readily controls dogs' right-left response differentiation, whereas dogs' go/no-go response differentiation i s controlled more readily by the qualitative than by the positional properties of such stimuli. It i s possible, therefore, that the relative control of pigeons' keypecking by food-access and light duration also may vary across discrimination tasks. For example, superior control by food-access duration may be specific to the go/no-go procedure in which one duration always serves as an S + for reinforcement and the other as an S -. The present experiments, therefore, were designed to replicate the above comparison using choice procedures in which reinforcement for a correct response was available after both durations. In Experiment 2A, food-deprived pigeons served under a discrete t r i a l * right-left choice procedure similar to that used in previous duration experiments (e.g., Church & Deluty, 1977; 32 Kinchla, 1970),. Trials began with either a short or a long duration of food-access or houselight presentation, followed by illumination of two side keys with green light. Pecks to the l e f t key were reinforced following the short duration of food access or light; right-key pecks were reinforced after the long duration of either stimulus. An elaboration of the delayed symbolic matching-to-sample procedure was used in Experiment 2Bi Here, pigeons were exposed to discrete t r i a l s that again began with one of two durations of food access or light. Following this, the two side keys were illuminated, one with red light and one with green, the right-l e f t position of red and green varying across t r i a l s . Reinforcement was provided for a peck to red following the short duration, and a peck to green following the longer duration of either stimulus. These procedures differed from the go/no-go paradigm of Experiment 1 in two important ways. Fi r s t , stimulus control was assessed by the accuracy of choice behavior rather than by differential response rates,. Second, both stimulus durations served as an S* for the availability of reinforcement. The procedures of Experiment 2A and Experiment 2B also differed from each other with respect to the response requirement; Experiment 2A involved a conditional right-left discrimination, whereas Experiment 2B involved a conditional color discrimination. A replication of the above results under these two procedures would provide convincing evidence for the generality of the differential effectiveness of the duration of food access and light as discriminative stimuli for pigeons. 33 EXPERIMENT 2A Method Subjects Two Silver King pigeons (Birds 12 and 13) served as the subjects* Bird 12 was experimentally naive; Bird 13 had served in a variety of previous experiments,. Both were maintained at approximately 85% of their free-feeding weights by the mixed grain obtained during experimental sessions and post-session feedings of maple peas* An unlimited supply of water and health grit was available in the individual home cages. Apparatus One wall of the BES-Foringer Model #PS-004 pigeon chamber contained a horizontal row of three clear plastic pecking keys that each reguired a force of 0.2 N to operate. An Industrial Electronics Engineers* Series 10 stimulus projector was mounted behind each key; these illuminated the keys with a uniform f i e l d of colored light. Centered below the keys was a Gerbrands Model #G5610 solenoid-operated feeder that permitted timed access to mixed grain. A 2.8 W lamp located within the hopper served to illuminate the grain presentations. The houselight consisted of two 2.8 W lamps hidden behind a plastic reflector above the keys; these provided a diffuse illumination of the chamber.. Control of experimental conditions and collection of data were performed by a Data General Nova 3 computer* 34 Procedure Preliminary Training At the end of preliminary training, both birds reliably ate from the raised illuminated grain feeder and pecked at both the l e f t - and fight-hand illuminated response keys. Choice Training, : Trials began with the presentation of either food access or houselight for a period of 5 sec or 10 sec. At the end of this period both side keys were illuminated with green light until one of these was pecked* If the correct key was pecked, a 5-sec grain reinforcer was presented, The left key was designated correct following 5 sec of food access or houselight; the right key was correct following 10^ -sec stimulus presentations* If the incorrect key was pecked (left after a 10-sec or right after a 5-sec stimulus presentation) the t r i a l was terminated and reinforcement was not presented* A correction procedure was employed; i f the pigeon made an incorrect response, the same i n i t i a l stimulus was presented again on the next t r i a l . A l l t r i a l s were separated by a 30-sec i n t e r t r i a l interval in which the chamber was dark. Each of the four i n i t i a l stimuli were presented five times within a session (correction t r i a l s excluded) in a randomly determined order* After 70 sessions with this procedure, the reinforcement schedule for correct responses was changed from continuous to partial reinforcement in preparation for subsequent testing (see Experiment 3). Por 20 sessions a 75% reinforcement schedule was in effect; correct responses were reinforced with a probability of .75, The remaining correct responses had the same outcome as 35 incorrect responses, except that correction t r i a l s did not occur* During the last 40 sessions for each bird, the reinforcement schedule for correct responses was decreased to 50%. Data Analysis Control by duration was assessed by choice accuracy, defined as the percentage of t r i a l s on which the correct side key was pecked. Separate accuracy scores were computed for food-access and houselight trials,. Accuracy scores were calculated for blocks cf 10 consecutive sessions and plotted as a function of training; higher scores reflected better control of choice behavior by stimulus duration,. Results Figure 7 shows both bird's accuracy scores on food-access and houselight t r i a l s as a function of training; Stimulus duration did acguire control of choice behavior, as reflected by the increase in accuracy scores to above the chance (50%) level. However, the extent of this control by duration differed for the two stimuli; food-access duration acguired more control (i.e., yielded higher accuracy scores) than did houselight duration* These results were consistent across the two birds and across the three reinforcement schedules. 36 Figure 4, Choice accuracy (percent correct) on food-access and houselight t r i a l s as a function of training for the two pigeons in Experiment 2A, In order. A, B, and C refer to the 100%, 75%, and 50% schedules of reinforcement for correct responses; the vertical dashed lines indicate the transition between these schedules. •-• FOOD ACCESS o—o HOUSELIGHT BLOCKS OF 10 SESSIONS 38 EXPERIMENT 2B Met hod -Subjects The subjects were White King pigeons 3 and 4 from Experiment 1, Silver King pigeon 12 from Experiment 2A, two additional Silver King pigeons (Birds 14 and 15), and one Homing pigeon (Bird 16), Birds 14, 15, and 16 each had served in a variety of previous experiments. A l l birds were maintained at approximately 85% of their free-feeding weights by grain obtained during experimental sessions and post-sessional feedings of maple peas. Water and health grit were freely available in the individual home cages* Apparatus The equipment used was the same as that described for Experiment 2A„ Procedure No preliminary training was necessary and a l l subjects were started immediately on the choice procedure* The choice procedure of this experiment involved a variation of the delayed symbolic matching-to-sample paradigm. The two-part t r i a l s began with the presentation of a sample stimulus for either a short or a long duration* Sample offset was followed immediately by illumination of the side keys with a red and a green comparison stimulus, the position of red and 39 green varying across t r i a l s . Pecks at the red comparison key after short samples, or at the green key after long samples produced a 5-sec grain reinforcer. Pecks at red after long samples or green after short samples terminated the illumination of both keys and initiated the 30-sec i n t e r t r i a l interval; these incorrect responses were always followed by a correction t r i a l . For a l l birds food access served as the sample on half of the t r i a l s ; on the remaining t r i a l s the sample was houselight for Birds 3, 4, and 12, and feeder light for Birds 14, 15, and 16, I n i t i a l l y the short and long samples were 5 sec and 10 sec, respectively, The short sample later was decreased to 2 sec in an attempt to improve matching performance in preparation for subsequent testing. 1- At the time of this change. Birds 3 and 4 had completed the experiment, and Birds 15 and 16 had not yet started. Birds 13 and 14 were in the course of training, and consequently received some sessions in which the short sample was 5 sec, followed by sessions in which the short sample was 2 sec, i In each session the four samples were presented five times each (excluding the correction t r i a l presentations) in a randomly determined order,. Birds 3, 4, 12, 14, 15, and 16 received 120, 120, 50, 110, 120, and 30 sessions, respectively* 1 An investigation pigeons' keypecking currently in progress* of by the effect of delays on control of food-access and light duration i s 40 Data Analysis Control by stimulus duration was assessed by the percentage of t r i a l s on which a correct response occurred (matching accuracy);. Separate matching accuracy scores were calculated for food-access and light trials,. For each, the scores were calculated for blocks of 10 consecutive sessions and plotted as a function of training, Results Figure 8 shows matching accuracy as a function of training for the six birds; Again, the same effects were revealed; The duration of both food-access and light samples controlled responses to the comparison stimuli; however, the control by food-access duration was superior to the control by either houselight duration ( Birds 3, 4, and 16) or feeder-light duration (Birds 12, 14, and 15). Discussion These experiments demonstrated the generality of the differences in control by food-access and light duration across discrimination procedures. The results of Experiment 1 were replicated in the present experiments with two choice procedures that differed in a number of ways from each other, and from the go/no-go procedure of Experiment 1. Food-access duration clearly was more effective than light duration in controlling pigeons* keypecking under a variety of conditions* 41 Figure 5. Matching accuracy (percent correct) on food-access and light t r i a l s as a function of training for the six pigeons in Experiment 2B. The vertical broken lines (Birds 3 and 4) indicate the transition from 10- and 5-sec samples to 10-and 2-sec samples,. •—• FOOD ACCESS o—o HOUSELIGHT •—• FOOD ACCESS FEEDER LIGHT BLOCKS OF 10 SESSIONS 43 EXPERIMENT 3 A d i s c r i m i n a t i o n between two s t i m u l i s e l e c t e d from a dimension does not always r e f l e c t s t i m u l u s c o n t r o l by t h a t dimension. For example, a d i s c r i m i n a t i o n between red and green may suggest c o n t r o l by wavelength; however, c o n t r o l by some other dimension, such as i n t e n s i t y may a l s o r e s u l t i n d i f f e r e n t i a l responding to red and green* One way to t e s t f o r c o n t r o l by a dimension a f t e r two-value d i s c r i m i n a t i o n t r a i n i n g i s to conduct a g e n e r a l i z a t i o n t e s t i n which new valu e s along the dimension are presented. I f the dimension c o n t r o l s behavior, responses t o the new s t i m u l i should vary as a s y s t e m a t i c f u n c t i o n of t h e i r p o s i t i o n on the continuum* The main purpose of t h i s experiment was to demonstrate stimulus c o n t r o l by the d u r a t i o n dimension, Such a demonstration was e s s e n t i a l to the present r e s e a r c h f o r two reasons* F i r s t , although previous experiments (e,g*, Reynolds & Cat a n i a , 1962; Stubbs, 1968) had demonstrated s t i m u l u s c o n t r o l by the d u r a t i o n dimension, such c o n t r o l had not been demonstrated with d u r a t i o n of food access* Second, because the present r e s e a r c h i n v o l v e d comparisons between s t i m u l i , i t had to be shown t h a t the same dimension c o n t r o l s responding i n both cases* The present study f o l l o w e d Experiment 2A i n which pigeons had r e c e i v e d r i g h t - l e f t c h o i c e d i s c r i m i n a t i o n t r a i n i n g with 5-sec and 10-sec d u r a t i o n s of food-access and h o u s e l i g h t * In the present experiment, g e n e r a l i z a t i o n t e s t i n g was conducted with i n t e r m e d i a t e d u r a t i o n s of these s t i m u l i , and the r e l a t i o n s h i p 44 between test duration and choice behavior was investigated. Method -Subjects and Apparatus The subjects and apparatus were those of Experiment 2A. Procedure This experiment directly followed the right-left choice training with 5-sec and 10-sec durations of food-access and houselight reported in Experiment 2A, The procedure used here was identical to that of Experiment 2A except that two types of t r i a l s were arranged. Choice training t r i a l s occurred with a probability of ,75, and reinforcement for correct choices on these t r i a l s (left after 5-sec and right after 10-sec stimulus presentations) occurred with a probability of *75. The remaining t r i a l s were generalization test t r i a l s * During test t r i a l s , either food access or houselight was presented for one of nine durations: 5,5, 6,0, 6,5, 7.Or 8,0, 8.5, 9f0# or 9,5 sec, On a given test t r i a l , each test duration of each stimulus (food access or light) occurred with a probability of 1/18. Following a test presentation, both side keys were illuminated and a peck to either key ended the t r i a l but did not produce reinforcement; The correction procedure remained in effect on training t r i a l s , but was not used during test t r i a l s . Generalization testing was conducted during 120 sessions by which time each test duration of both food access and houselight had been presented at least 25 times. 45 Data Analysis For each of the 120 test sessions, a record was kept of the number of times that the right and l e f t keys were pecked following each test duration of the stimuli. Data from food-access and houselight test t r i a l s were analysed separately.„ For each, a function relating the proportion of left-key responses (correct for 5-sec stimulus presentations) to test durations of the stimuli was plotted. Each point was the ratio of the number of left-key responses at a test-stimulus duration divided by the total number of presentations of that test-stimulus duration. Results The functions relating proportion of left-key responses to test-stimulus durations are presented in Figure 6* The left-key responses of both birds clearly decreased as a function of increases in the durations of the stimuli. No major differences between the functions obtained on food-access t r i a l s and those obtained on houselight t r i a l s are apparent* Discussion This experiment clearly demonstrated stimulus control by the duration dimension;. Choice behavior varied as a function of the test durations of both the food-access and light stimuli. Thus, the difference in control by the training durations of food access and houselight was not due to different controlling dimensions. The present results are consistent with several other 46 Figure 6. Proportion of left-key responses as a function of stimulus duration for the two pigeons in Experiment 3, Each point represents data from at least 25 t r i a l s * FOOD ACCESS HOUSELIGHT STIMULUS DURATION (SEC) 48 demonstrations of control by the duration dimension. For example, Reynolds and Catania (1962) reinforced pigeons1 keypecks after one dark-key duration (S+) but not after several other dark-key durations (S~). Keypecking rates after the S-durations varied as a function of the difference between the S -and the S+ durations* Stubbs (1968) reported similar results with a choice procedure in which one response was reinforced following any of a class of short keylight durations, and another response was reinforced following any of a class of long durations* Accuracy of responding was a direct function of the relative difference in time between a particular keylight duration and the cutoff between these two classes, Church and Deluty (1977) demonstrated control of rats' choice behavior by the duration dimension, using the same technique as used in the present experiment. Rats were given right-left choice training with two blackout durations, followed by generalization tests with intermediate durations,. Choice responses appropriate to the short duration decreased as a function of increases in the intermediate durations, The present experiment extends these studies by demonstrating similar control by the duration of food access. GENERAL DISCUSSION These experiments clearly demonstrated that food-access duration i s superior to light duration in controlling pigeons' operant keypecking. This difference in control is reliable, and generalizes to a number of experimental conditions. In Experiments 1 and 2, every pigeon discriminated durations of 49 food access better than durations of light; these results were not specific to the particular kind of light, to the reinforcer, or to the operant discrimination procedure employed, Experiment 3 demonstrated that the keypecking was controlled by the duration dimension on both food-access and light t r i a l s . As discussed in the Introduction, there were three major rationales for the present research* The results will be discussed with specific reference to these three topics in the following sections. The Effectiveness of Food in Stimulus Control of Operant Behavior The results of this research are consistent with the assumption that food i s very effective in controlling behavior. Previous evidence for this assumption came from studies that demonstrated that animals often can remember certain properties of food over long intervals. The present research extends these findings by demonstrating^ that antecedent control by food developed faster and to a greater degree than control by light. Any of a number of properties may be responsible for the superior effectiveness of food in the present experiments. It i s possible, for example, that the pigeons' behavior during the food-access presentations mediated the control by duration. There is considerable evidence that animals are sensitive to several aspects of their behavior. Beninger, Kendall, and Vanderwolf (1974) demonstrated that rats could discriminate between different classes of their own behavior (e,g» , rearing, face-washing, etc). Other studies have demonstrated that 50 pigeons can discriminate the number of their keypecks (e.g., B i l l i n g , 1967; Wilkie, Webster, & Leader, 1979), as well as the time between keypecks (Reynolds, 1966), Apparently animals are also sensitive to the duration of their responses,. Piatt, Kuch and Bitsgood (1973) reported that rats learned to emit lever presses of a certain range of durations, and Ziriax and Silberberg (1978) found that pigeons could discriminate between keypecks that differed only by a few milliseconds. In view of these findings one might expect that pigeons would readily discriminate their eating times, Because the duration of eating during food access likely was correlated with the duration of the food-access presentations, a discrimination of eating times might effectively serve to mediate the duration discrimination. On the other hand, the duration of orienting or other behaviors that occurred during light may not have been as closely correlated with the duration of the light presentations, or as effective in mediating the duration discrimination. Attentional mechanisms might also account for the superior effectiveness of food in controlling pigeons' behavior. For example, the rewarding value and/or the response-eliciting properties of food may ensure attention to the stimulus on food-access t r i a l s ; whereas, the pigeons may not always have attended to the light presentations for their entire duration, or on some t r i a l s . Attention to the stimulus conceivably could affect both the speed of learning about the relevent dimension, as well as performance on the task once the relevent dimension has acquired control* The complexity of the food-access stimulus may also have 51 contributed to i t s effectiveness. Food presentations consisted of several elements, including the noise associated with raising and lowering of the grain hopper and the sight of grain. Perhaps the greater number of stimulus elements present on food-access t r i a l s increased their salience. The superior control by food also could be related to the dimension chosen to compare food and light . For example, theories of selective attention (e.g.. Mackintosh, 1974) have suggested that certain dimensions of a stimulus may be more salient than other dimensions and control by the salient dimensions will sometimes interfere with, or overshadow control by the relevent dimension, It i s possible that duration i s a particularily salient property of food but not of light. The duration of access to food likely i s of particular importance to a hungry animal;. If other dimensions of light (e>g;, intensity) are more salient than i t s duration, the poorer control by light could be due to greater overshadowing by these nonrelevent dimensions* Finally, i t i s possible that the superior control by food access reflects i t s greater biological importance to the animal. For example, proponents of the recent biological constraints on learning position (e.g., Seligman, 1970; Shettleworth, 1972) have suggested that animals are predisposed to learn about certain events more readily than others because of the adaptive significance of such learning in nature. The present experiments found that control by food access, a presumed biologically important stimulus, was superior to control by light, a presumed more \"neutral\" event. Although this finding 52 is consistent with the constraints on learning viewpoint, considerable additional research would be required to demonstrate that animals are predisposed to learn about biologically important events, as- a class^ more readily than about more \"neutral\" events. Comparisons Among Stimuli The present results provide direct empirical evidence for the assumption that stimuli differ in their a b i l i t y to control operant behavior* The difference in control of pigeons' behavior by food and light was not confounded with differences in the discriminability of the dimensions that controlled the behaviors because: 1) the same values from a dimension that was common to both stimuli served as the basis of the comparison and, 2) evidence was provided that this dimension of both stimui controlled the behaviors,. In addition to demonstrating such differences between stimuli, this research also illustrates the suitability of stimulus duration as a dimension with which to compare stimuli. Comparisons of stimuli within a duration paradigm offer several advantages over other procedures for comparing stimuli* F i r s t , using duration as the basis for such comparisons avoids the necessity of scaling values of different dimensions. Second, the values of the duration dimension can be accurately specified and controlled without the use of sophisticated procedures such as are needed for specification and control of values from dimensions such as intensity or wavelength. Third, because duration i s a property of a l l stimuli* comarisons among any 53 number of s t i m u l i can be conducted within a duration paradigm. This procedure for comparing st i m u l i could be applied to several areas of investigation within the stimulus control l i t e r a t u r e . One such area i s the recent interest i n of stimulus-response and stimulus-reinforcer s p e c i f i c i t i e s (e.g., Shettleworth, 1972). For example, some investigators have reported that v i s u a l stimuli are more e f f e c t i v e than auditory s t i m u l i i n c o n t r o l l i n g pigeons* food-reinforced behaviors, whereas control by the auditory s t i m u l i i s superior i n aversive conditioning or avoidance tasks (e.g., Delius & Emmerton, 1978; Foree & LoLordo, 1974), Foree and LoLordo suggested that such res u l t s are consonant with the knowledge that, i n nature, pigeons f i n d food by sight, but r e l y on auditory cues when predators approach, However, i n these studies the dimensions of the s t i m u l i that controlled behavior were not i d e n t i f i e d , and no attempts were made to eguate the values of such dimensions for d i s c r i m i n a b i l i t y , Thus, i t i s not known whether auditory and visual s t i m u l i per se are d i f f e r e n t i a l l y e f f e c t i v e i n appetitive and aversive s i t u a t i o n s , or whether pigeons* behavior i s controlled by d i f f e r e n t stimulus dimensions i n the two situations* Such ambiguity could be prevented by using a common dimension to compare the s t i m u l i . For example, by using the duration dimension as a basis for comparing a vis u a l and an auditory stimulus under appetitive and avoidance tasks, one could determine whether the r e l a t i v e effectiveness of the s t i m u l i themselves i s s p e c i f i c to the s i t u a t i o n . . 54 The Bole of the Stimulus in Control by Duration The present research demonstrated that the stimulus one selects to study duration discrimination can determine the speed of acquisition as well as the accuracy of such discriminations. This finding has certain implications for the study of temporal discriminations in animals* For example, an animal's accuracy in discriminating the duration of a given stimulus may not accurately reflect the animal's timing a b i l i t i e s , because how accurately the animal discriminates event duration depends in part on the nature of the event* The present results are relevent to a recent theory of temporal discrimination in animals (Church, Getty, & Lerner, 1976). This theory proposed three factors that limit an animal's accuracy in estimating temporal intervals: 1) inattention to the signal on some t r i a l s ; 2) variability in starting to time the duration when the signal begins and/or in stopping to time the duration when the signal ends; and 3) factors related to the signal duration i t s e l f . The present results, which demonstrated that pigeons' accuracy in discriminating durations i s limited by the nature of the stimulus, could be interpreted in terms of either of the f i r s t two factors proposed by this theory. For example, the presence of food on food-access t r i a l s may have ensured attention to the signal on a l l such t r i a l s and/or decreased the variability in starting and^stopping to time the duration. The present research may also have implications for theories of timing mechanisms in animals* Roberts and Church (1978), for example, recently proposed an elaborate theory of 55 internal clock control to describe timing processes in the rat. This theory suggests that rats possess some sort of internal clock, similar in many repects to the familiar stopwatch, that functions to measure durations. One explicit assumption of this theory is that the rats' clock, like a stopwatch, measures time independently of modality* as evidence for this assumption, Roberts and Church demonstrated that rats were able to add time spent in light to time spent in sound and respond on the basis of the sum of these times. This was interpreted as evidence that the rats timed light and sound at the same rate* The present findings suggest that i t i s important to explore the generality of such qualitative aspects of control by duration over a wider range of stimuli. For example, although rats may measure time independently of stimulus modality, i t is conceivable that their measurement of time i s not independent of other properties of the stimulus. It would be interesting to investigate whether rats* timing of light and sound i s the same as their timing of a stimulus such as food that: 1) i s a more complex stimulus, 2) likely controls a specific behavior of the organism, and 3) has greater \"biological importance\" for the organism* Reference Note Spetch, H, L. , & Wilkie, D, Mi Pigeons' keypecking topographies under a schedule of alternating food and water rewards* Paper presented at the meeting of the Canadian Psychological Association, Quebec City, June 1979. 57 Eeferences Beninger, R. J , , K e n d a l l , S, B., & Vanderwolf, C* _H, The a b i l i t y of r a t s to d i s c r i m i n a t e t h e i r own behaviours. Canadian J o u r n a l of Psychology, 1974, 28, 79-91. Blough, D. S, Delayed matching i n the pigeon. J o u r n a l of the Experimental A n a l y s i s of Behavi o r , 1959, 2, 151-160. B o t j e r , S. W. , & Hearst, E. Food d e l i v e r y as a c o n d i t i o n a l s t i m u l u s : Feature l e a r n i n g and memory i n pigeons. J o u r n a l of the Experimental A n a l y s i s of Behavior, 1979, 31, 189-207. C a p a l d i , E, J . A s e q u e n t i a l h y p othesis of i n s t r u m e n t a l l e a r n i n g * In W. K. Spence, & J . T* Spence (Eds*), The - Psychology o f l e a r n i n g and m o t i v a t i o n , (Vol.1)* New York: Academic P r e s s , 1967. C a p a l d i , E. J . , & Spivey, J , E. I n t e r t r i a l r e i n f o r c e m e n t and a f t e r e f f e c t s at 24-hour i n t e r v a l s * Psychonomic S c i e n c e , 1964, X, 181-182* C a r t e r , D. E- , £ Eckerman* D, A. Symbolic matching by pigeons: Rate of l e a r n i n g complex d i s c r i m i n a t i o n s p r e d i c t e d from simple d i s c r i m i n a t i o n s * S c i e n c e , 1975, 187, 662-665. Chow, K. L. Stimulus c h a r a c t e r i s t i c s and r a t e of l e a r n i n g v i s u a l d i s c r i m i n a t i o n s by e x p e r i m e n t a l l y naive monkeys* American J o u r n a l o f Psychology, 1953, 66, 278-282, Church, fi. M., & Del u t y , M. Z, B i s e c t i o n of temporal i n t e r v a l s . J o u r n a l of Experimental Psychology: Animal Behavior Processes, 1977, 3, 216-228* Church, R, W . , Getty, D, J , , & Lerner, N, D* Duration d i s c r i m i n a t i o n by r a t s , J o u r n a l of Experimental Psychology: Animal Behavior Processes, 1976, 2, 303-312. Cruse, D, B., V i t u l l i , W., & Dertke, M* D i s c r i m i n a t i v e and r e i n f o r c i n g p r o p e r t i e s o f two types of food p e l l e t s , J o u r n a l o f the Experimental A n a l y s i s of Behavior, 1966, 9, 293-303, D e l i u s , J . D. & Emmerton, J . Stimulus-dependent asymmetry i n c l a s s i c a l and i n s t r u m e n t a l d i s c r i m i n a t i o n l e a r n i n g by pigeons* The P s y c h o l o g i c a l Record, 1978, 28, 425-434*_ 58 Dobrecka, C , Szwejkowska, G., & Konorski, J* Q u a l i t a t i v e . v e r s u s d i r e c t i o n a l cues i n two forms o f d i f f e r e n t i a t i o n * S c i e n c e , 1966, 153, 87-89* Elsmore, T. F. C o n t r o l of responding by s t i m u l u s d u r a t i o n * J o u r n a l of - the Experimental A n a l y s i s of Behavior, 1971, 16, 81-87,. Elsmore, T, F. Duration d i s c r i m i n a t i o n : E f f e c t s of p r o b a b i l i t y of s t i m u l u s p r e s e n t a t i o n * J o u r n a l of the Experimental A n a l y s i s o f Behavior, 1972, J 8 , 465-469. Foree, D. D,, & LoLordo, V. M, A t t e n t i o n i n the pigeon: D i f f e r e n t i a l e f f e c t s of f o o d - g e t t i n g versus shock-avoidance procedures. J o u r n a l of Comparative r and P h y s i o l o g i c a l Psychology. 1975, 85, 551-558. Green, D, M. , & Swets, J* A. S i g n a l d e t e c t i o n theory and psychophysics. New York: Wiley, 1966, H a r r i s o n , J , W . , & B r i g g s , B. O r i e n t a t i o n and l e v e r responding i n a u d i t o r y d i s c r i m i n a t i o n s i n s g u i r r e l monkeys* J o u r n a l o f the Experimental A n a l y s i s of Behavior, 1977, 28, 233-241; Hearst, E, Delayed a l t e r n a t i o n i n the pigeon* J o u r n a l of the Experimental A n a l y s i s of Behavior, 1962, 5, 225^228. Hersen, M, 6 Barlow., D. H. S i n g l e - c a s e experimental - d e s i g n s : S t r a t e g i e s f o r s t u d y i n g behavior change;. New York: Pergamon Pre s s , 1976. Honig, W. K*, S James, P. H* . B. (Eds,*)* Animal memory* New York: Academic Press, 1971,.. Hulse, S., Fowler, H., & Honig, W» K* (Eds.).. C o g n i t i v e a s p e c t s of animal behavior* H i l l s d a l e , N. J * : Lawrence Erlbaum A s s o c i a t e s , 1978, J a r v i k , B* E, D i s c r i m i n a t i o n of c o l o r e d food and food s i g n s by primates. J o u r n a l of Comparative and P h y s i o l o g i c a l Psychology, 1953, 46, 390-392* K i n c h l a , J . D i s c r i m i n a t i o n of two a u d i t o r y d u r a t i o n s by pigeons. P e r c e p t i o n and Psychophysics, 1970, 8, 299-307. Lawicka, W, The r o l e of s t i m u l u s modality i n s u c c e s s i v e d i s c r i m i n a t i o n l e a r n i n g and d i f f e r e n t i a t i o n l e a r n i n g . B u l l e t i n of the P o l i s h Academy of S c i e n c e s , 1964, 1_2, 35-38, 59 Lydersen, T, , Perkins, D*, & Chairez, H,; Effects of fixed-ratio sample and choice requirements upon oddity matching. Journal of the Experimental Analysis of Behavior, 1977, 25!, 97-10 1. Mackintosh, N. J. The psychology of animal learning* New York: Academic Press, 1974, Mackintosh, N, J. Stimulus control: Attentional factors* In W, K* Honig, and J. Et R, Staddon (Eds), Handbook of Operant Behavior* New Jersey: Prentice-Hall, 1977i „ Maki, W* S. , Jr., Moe, J. C*, & Bierley, C, M, Short-term memory for stimuli, responses, and reinforcers. Journal of Experimental Psychology: Animal Behavior Processes, 1977, 3, 156-177, Medin, D. L., Roberts, W. A. , & Davis, R* T, Processes of animal memory, Hillsdale, N. J. : Lawrence Erlbaum Associates, 1976. Perikel, J, J,, Richelle, M*, & Maurissen, J, Control of keypecking by the duration of a visual stimulus. Journal of the Experimental Analysis of Behavior, 1974, 22, 131-134; Piatt, J. R,, Kuch, D, 0., & Bitsgood, S, C. Rats* lever-press duration as psychophysical judgements of time. Journal of the Experimental Analysis of Behavior; 1973, J_9, 239-250; Pschirrer, M, Goal events as discriminative stimuli over extended i n t e r t r i a l intervals;; Journal of Experimental Psychology, 1972, 96, 425-432,; Reynolds, G, S. Discrimination and emission of temporal intervals by pigeons* Journal of the Experimental Analysis of Behavior, 1966, 9, 65-68, Reynolds, G. S,, & Catania, A* C* Temporal discrimination in pigeons* Science, 1962, 135, 314-315. Ri l l i n g , M, Number of responses as a stimulus in fixed interval and fixed ratio schedules; Journal of Comparative and Physiological Psychology, 1967, 63, 60-65, Roberts, S,, & Church, R. M. Control of an internal clock; Journal of Experimental Psychology: Animal Behavior Processes, 1978, 4, 318-337, Seligman, M, E. P, On the generality of the laws of learning* Psychological Review, 1970, 77, 406-418,: 60 Shettleworth, S* J. Constraints on learning* advances in the Study of Behavior, 1 9 7 2 , 4 , 1-681 \" r _ Sidman, ft.. Tactics of scientific research,. New York: Basic Books, Inc., 1960;. Skinner, B. F. The behavior of organisms* New York: Appleton-Century-Crofts, 1 9 3 8 . Staddon, J. E. E. Temporal control, attention* and memory* Psychological Beview, 1 9 7 4 , 81., 3 7 5 - 3 9 1 . Stubbs, a . The discrimination of stimulus duration by pigeons* Journal of the Experimental Analysis of Behavior, 1 9 6 8 , 1 1 , Stubbs, D„ A. Scaling of stimulus duration by pigeons* Journal of the Experimental analysis of Behavior, 1 9 7 6 , 2 6 , 1 5 - 2 5 * Stubbs, D, a , Response bias and the discrimination of stimulus duration* Journal of the Experimental analysis of Behavior, 1 9 7 6 , J 5 , 2 4 3 - 2 5 0 * Wall, a , M,, & Goodrich, K. P. Differential responding on reinforcement and nonreinforcement t r i a l s occurring in fixed repeated patterns* Psychonomic Science, 1 9 6 4 , _1, 1 9 3 -194,. Wilkie, D. M. Delayed symbolic matching to sample in the pigeon* The Psychological Becord, 1 9 7 8 , 2 8 , 4 6 3 - 4 6 9 . Wilkie, D. M,, S Masson, M. E,. attention in the pigeon: a reevaluation* Journal of the Experimental - analysis of Behavior, 1 9 7 6 , 2 6 , 2 0 7 - 2 1 2 . Wilkie, D. H, , Webster, J..B,, & Leader, L, . G, Unconfounding time and number discrimination in a Mechner counting schedule* Bulletin of the Psychonomic Society, 1 9 7 9 , 1 3 , 3 9 0 - 3 9 2 , Wilkie, D, M., & Wilson, C, S, Control of pigeons' keypecking by trace stimuli* Journal of the : Experimental analysis of Behavior, 1 9 7 7 , 2 7 , 2 9 3 - 2 9 9 i Williams, B. a , Color alternation learning in the pigeon under fixed-ratio schedules of reinforcement* Journal of the Experimental analysis of Behavior, 1 9 7 1 , 1_5_, 1 2 9 - 1 4 0 . Ziriax, J. fl,, & Silberberg, a . Discrimination and emission of different key-peck durations in the pigeon,. Journal of Experimental Psychology: animal Behavior Processes, 1 9 7 8 , 4 , 1 - 2 1 . 2 2 3 - 2 3 8 , "@en ; edm:hasType "Thesis/Dissertation"@en ; edm:isShownAt "10.14288/1.0100233"@en ; dcterms:language "eng"@en ; ns0:degreeDiscipline "Psychology"@en ; edm:provider "Vancouver : University of British Columbia Library"@en ; dcterms:publisher "University of British Columbia"@en ; dcterms:rights "For non-commercial purposes only, such as research, private study and education. Additional conditions apply, see Terms of Use https://open.library.ubc.ca/terms_of_use."@en ; ns0:scholarLevel "Graduate"@en ; dcterms:title "Pigeons discriminate durations of food access better than durations of light"@en ; dcterms:type "Text"@en ; ns0:identifierURI "http://hdl.handle.net/2429/21674"@en .