Printer Friendly

Effects of reinforcement context on initial link responding in concurrent chain reinforcement schedules.

Choice researchers have shown that preference can be affected by several different variables, such as reinforcer magnitude and rate (see Mazur & Fantino, 2014). Choice researchers employ a concurrent-chains procedure which separates the choice stage (initial link) from the outcome stage (terminal link), where primary reinforcement is obtained (e.g., food for a hungry pigeon). Many studies have consistently shown that the relative duration (inverse of reinforcer rate) of each link affects preference. The initial-link effect refers to a decrease in preference for the richer outcome as the overall duration of the initial link increases (see Fantino, Preston & Dunn, 1993, for review). For example, if the average initial link duration increases from concurrent variable interval (VI) 15 s schedules to concurrent VI 150 s schedules, preference for the initial link correlated with the richer terminal link decreases. Likewise, the terminal-link effect refers to a decrease in preference for the richer terminal link as the average terminal link duration increases, so long as the duration ratio between terminal link outcomes is constant (MacEwen, 1972; Grace, 1994; Grace & Bragason, 2004; Williams & Fantino, 1978).

Taken together, both the initial- and terminal-link effects suggest that the relative duration of each link of the concurrent chain schedule affects choice. However, research on relative initial and terminal duration on choice has primarily been limited to factors operating within those initial or terminal links (i.e., local context). Few studies have tested whether factors known to affect choice, such as added reinforcement and changes in relative duration between choice opportunities, also systematically affect choice when the initial- and terminal-links are not present (i.e., global context). In one study, Goldshmidt et al. (1998) presented conditioned reinforcers during the inter-trial interval (ITI) of a concurrent chains procedure. Results indicated that neither providing the same conditioned reinforcers used for the richer terminal link during the ITI, providing the same conditioned reinforcers used for the initial links during the ITI, nor providing response-independent food during the ITI significantly changed choice proportions during a concurrent chains procedure. Although the results supported a local contextual view of choice (i.e., added reinforcement outside of the concurrent chains procedure did not affect choice), there were no response-reinforcer contingencies during any of the conditions. This resulted in low response rates during the ITI for both of the keylight manipulations (see Experiments 2 & 3; Goldshmidt et al., 1998).

Similar to Goldshmidt et al., Williams and Fantino (1996) used a response-dependent fixed interval (FI 200 s) pre-choice period that led directly to the subsequent initial link. Again, the data failed to show a difference in choice proportions between conditions with or without the pre-choice FI period. However, similar to Goldshmidt et al. (1998), none of the experiments contained response-dependent unconditioned reinforcement during the FI pre-choice period, and the FI pre-choice period again produced low response rates. Therefore, it is still unknown if unconditioned reinforcement that occasions significant responding during periods when neither concurrent chains link is present will affect initial link choice proportions.

Outside of directly assessing changes in choice proportions, these previous researchers employed procedures conceptually similar to multiple schedules, where one component (i.e., concurrent chain) alternated with a second component (i.e., added reinforcement context). Holding reinforcement rate constant for the first component while decreasing reinforcement rate for the second component typically results in increased response rates during the first component, even though reinforcement rates during the first component has not changed. This is called behavior contrast (Reynolds, 1961; Williams, 2002). However, when reinforcement in the second component is response-independent, behavioral contrast occurs less reliably (Halliday & Boakes, 1971; Thompson & Corr, 1974; Weisman & Ramsden, 1973). Thus, similar to contrast, using response-independent reinforcement schedules may have decreased the researchers' ability to measure a global context effect on response rates and, thus, choice proportions.

Behavior contrast has been previously demonstrated within the terminal links of concurrent chain schedules (Hursh & Fantino, 1973; LaFiette & Fantino, 1989). In both cases, when the duration of one of the terminal links was increased, the response rate for the unchanged terminal link increased. However, like much of the previous research, this contrast effect was brought about by changes to the terminal link. The experiment by LaFiette and Fantino (1989) also compared reinforcement context between an open and closed economy, where subjects could either acquire some (open) or all (closed) of their food through the experimental procedure. This would also constitute a test between a local and global context, as the open economy provided free food (unconditioned reinforcement) after each session, outside of the presence of either concurrent chains link. Like the previous experiments there was no choice proportion difference between conditions. This is consistent with the previous studies (Goldshmidt et al., 1998; Williams & Fantino, 1996), as there was no response requirement during the open economy outside of the concurrent-chains procedure. That is, the pigeon was fed after the session regardless of responding during the session to maintain a minimum body weight.

The present study tested the effects of alternating either a response-dependent or response-independent reinforcement between blocks of concurrent chains reinforcement schedules on initial link choice proportions. The response-dependent and independent reinforcement consisted of FI and fixed time (FT) reinforcement schedules, respectively. Therefore, unlike Goldshmidt et al. (1998), LaFiette and Fantino (1989), and Williams & Fantino (1996), response-dependent food was presented during FI trials. In addition, these reinforcement schedules would result in a meaningful range of response rates during the added reinforcement schedules. We hypothesized that adding reinforcement schedules outside of the concurrent-chains procedure would have two effects. First, in terms of choice proportions we hypothesized that adding reinforcement between concurrent chains presentations would function similarly to the initial-link effect. That is, initial link choice proportions would decrease as added reinforcement trial duration increases (i.e., reinforcement rate decreases). Second, consistent with behavior contrast research, concurrent chain response rates would increase as the added reinforcement rate decreased. In addition, the global context effect on response rates would be greater with FI, relative to FT schedules, which would influence choice proportions during the added FI reinforcement schedule more than the added FT reinforcement schedule.

Materials and Methods

Subjects

Twelve white carneux pigeons (Columba livia) with extensive operant experience served as subjects. Pigeon KD963 was excluded from analysis due to chamber malfunction. Pigeons were weighed daily and kept at approximately 80 % of their free feeding weight (FFW) through supplemental feeding. Pigeons were withheld from testing if they were > 85 % FFW, or < 75 % FFW. Pigeons were kept on a 14:10 light/dark cycle and had free access to both water and grit.

Materials

Training and testing was conducted using 12 operant chambers (BRS-LVE model SEC 9381-D; 35.0 cm height, 30.9 width, 35.2 cm depth), each equipped with three visual stimuli projectors (combination of Med Associates and Industrial Electronics Engineers brand), three round (2.5 cm circumference) response keys mounted 25 cm above the floor grate, one houselight (centered horizontally above the center key, 3.3 cm below the chamber ceiling), and a square hopper (5.7 cm width, 5.0 cm height) centered 12.2 cm above the floor grate that provided grain as unconditioned reinforcement. Keylights were aligned horizontally with the middle key centered on the intelligence panel and each side key splayed 8.3 cm on either side of the middle key, center-to-center. A force of 0.15 N displaced the keylight and closed a circuit to register a response (i.e., keypeck). Twelve Dell Optiplex GX520 desktop PCs operating National Instruments Lab VIEW 8.0 software were used, which controlled the procedural parameters and recorded response data.

Concurrent Chains

Initial links began with the presentation of identical white stimuli signaling independent VI60 s reinforcement schedules on the left and right keylights. VI interval distribution was determined by an algorithm that generated 30 intervals (Fleshier & Hoffman, 1962). Intervals were randomly selected without replacement. The first response to either of the illuminated keys after either of the intervals elapsed darkened both keylights and illuminated the terminal link keylight corresponding to the side where the last response was made. Depending on the side of the last response, the terminal link was either a VI 15 s (red keylight; richer) or VI 45 s (green keylight; leaner) schedule. Terminal link schedule completion resulted in 2 s access to food. Side orientations for the rich and lean schedules switched after each condition after stability was reached (see below). A 2 s changeover delay operated during the initial links to decrease the likelihood of rapid alternation between the two initial links (Shull & Pliskoff, 1967).

Context Schedules

Either a FI or FT reinforcement schedule operated on the center key between concurrent chains presentations. For simplicity, the FI and FT schedules were considered context schedules in that they operated during the same session (except for baseline) as the concurrent chains procedure, yet there was no contingency between responding during the concurrent chains procedure and FI or FT context schedules. Each context schedule had three possible reinforcer schedule values: 20, 60, and 180 s. FI reinforcement schedules required one response to occur to the illuminated keylight after the delay had elapsed. There was no response-reinforcer requirement for FT schedules. An illuminated yellow keylight signaled FI schedules, whereas an illuminated white triangle with darkened background signaled FT schedules.

Experimental Design

Concurrent chains and context schedules never occurred at the same time; they alternated with a 1 s inter-trial interval (ITI). All illumination (keylights and houselight) was removed during the ITI. Sessions started with four consecutive context schedule trials followed by four consecutive concurrent chains trials. Each trial ended in food reinforcement. Blocks of four consecutive context schedule trials or concurrent chains trials alternated for approximately one hour. Total trials per session were adjusted to keep the overall session duration approximately constant. Pigeons completed a total of 48, 32, and 24 trials per session, for the context schedules corresponding to 20, 60, and 180 s, respectively. Baseline conditions contained no context schedules, contained a 1 s ITI, and lasted 32 trials.

We used a within-subjects design. Pigeons completed eight conditions: two baselines and three context schedule conditions for both FI and FT schedules. Condition order was randomized, except baseline. Baseline conditions only occurred during the first and last three scheduled conditions of the experiment (see Table 1, 2nd column).

Stability Criterion

The main dependent variable was initial link choice proportion during the concurrent chains procedure, and was calculated by dividing the total number of responses to the initial link correlated with the richer terminal link schedule (i.e., VI 15 s) by the total number of responses during both initial links for each session. Choice proportions were considered stable when: (1) each choice proportion over the last five sessions of the condition was not the highest or lowest for that condition, (2) each choice proportion over the last 5 sessions was within 10 % of each other, and (3) each condition was conducted for a minimum of 20 sessions, and a maximum of 35 sessions. However, no pigeon required more than 35 sessions to reach the initial link stability criterion (see Table 1).

Data Analysis

Analyzed data represented the mean choice proportion and response rates calculated from the last five sessions of each condition. Data were normally distributed (skewness between -0.8 and 0.8; kurtosis between -2 and 2). Therefore, we used 2x4 (response contingency [FI, FT] x context schedule duration [baseline, 20, 60,180]) repeated measures analysis of variance (ANOVA) to test differences in initial link choice proportions and response rates during the initial and terminal links, and context schedules. Baseline was entered as a 0 s context duration for statistical analyses. Significant main effects were followed up with two-tailed paired Student's t-tests to determine which comparison was significantly different. To correct for multiple post-hoc comparisons, the alpha threshold for each post-hoc measure was adjusted to 0.008.

Results

Table 1 shows choice proportion and response rate data for each pigeon across all conditions. Pigeons 17 and EA1752 both had stable choice proportions below 0.50 (preference for leaner terminal link alternative) during at least two conditions, and were excluded from subsequent analyses. (1) Figure 1 shows mean initial link choice proportions generally decreased with increasing context schedule duration, similar to the initial-link effect. However, these decreases were small--the median choice proportion change was 0.01-0.11 across context schedule conditions. In addition, baseline choice proportions were variable across pigeons. A repeated measures ANOVA showed a significant main effect for context schedule duration F(3, 24) = 6.23, p = 0.04, [[eta].sup.2] = 0.70; choice proportions generally decreased as the context schedule duration increased. Post-hoc paired t-tests showed only one significant choice proportion difference as a function of context schedule duration. Baseline choice proportions were greater than 180 s (p = 0.006; d = 0.61) context schedule durations. All remaining reported comparisons were only at the trend level. Choice proportions during the 20 s (p - 0.02; d = 0.69) and the 60 s (p = 0.04; d = 0.04) context schedule duration were generally greater than the 180 s context schedule duration. There was also a trend difference for greater 20 s context schedule duration choice proportions relative to the 60 s context schedule duration choice proportions (p = 0.06; d = 0.61). There was no main effect for response contingency F(1, 8) = 0.12, p = 0.74, and no interaction between context schedule duration and response contingency F(3, 24) = 3.38, p = 0.09.

Figure 2 shows the mean initial link response rates for each pigeon across all conditions. Initial link response rates generally increased with increasing context schedule durations. This was confirmed with a repeated measures ANOVA that showed a significant main effect for context schedule duration F(3, 24) = 8.45, p = 0.001; [[eta].sup.2] = 0.51. Post-hoc paired t-tests showed significant differences for two comparisons. 20 s context schedule duration initial link response rates were lower than both the 60 s (p = 0.001; d= 0.62) and 180 s (p = 0.007; d=0.74) context schedule initial link response rates. All remaining comparisons were at the trend level. Baseline initial link response rates were generally lower than the 180 s context schedule duration (p = 0.02; d = 0.57). Baseline initial link response rates were generally lower relative to the 60 s context schedule duration (p = 0.06; d = 0.42). There was no main effect for response contingency F(l, 8) = 0.82, p- 0.39, nor an interaction F(3, 24) = 0.14, p- 0.72. There were also no main effects nor an interaction for terminal link response rates (Table 1; sum of eighth and ninth columns; all p values >0.05).

Context schedule response rates (Table 1, rightmost column) were generally consistent with scheduled reinforcement rates. A 2 x 3 (response contingency [FI, FT] x context schedule duration [20, 60, 180]) repeated measures ANOVA tested for differences in context schedule response rates. There was a significant main effect for response contingency F(1, 8) = 42.62, p < 0.001, [[eta].sup.2] = 0.84, and main effect for context schedule duration F(2, 16) = 6.97, p = 0.007, [[eta].sup.2] = 0.47. As expected, FT responses rates were lower than FI response rates. Also, response rates generally decreased as reinforcement rates decreased during the context schedules. Post-hoc Student's t-tests showed a significant difference between the 20 s and 180 s (p = 0.008, d = 0.71) context schedule response rates. The remaining comparisons between 60 s and 180 s context schedule duration response rates (p - 0.04; d=0.28), and between the 20 s and 60 s context schedule duration response rates (p = 0.06; d = 0.47) were at the trend level. Thus, pigeons appeared sensitive to changing context schedule durations. There was no interaction between response contingency and context schedule duration F(2, 16) = 1.09, p = 0.36.

Discussion

Consistent with our initial-link effect hypothesis, choice proportions generally decreased as context schedule duration increased (reinforcement rate decreased). However, response-dependent (FI) and response-independent (FT) schedules of reinforcement did not differentially change choice proportions. Initial link response rate changes were consistent with behavior contrast (Williams, 2002), previous studies using concurrent chain schedules (Fantino & Davison, 1983; LaFiette & Fantino, 1989; Williams & Fantino, 1996), and our general response rate hypothesis. However, these results were inconsistent with our response rate context hypothesis and previous research (Goldshmidt et al., 1998; Williams & Fantino, 1996) in that initial link response rates did not differentially change during FI and FT contexts, similar to the choice proportions. Furthermore, contrast-like response rate effects were limited to the initial links.

The global context effect on choice proportions was not a function of response-reinforcer contingencies. The average choice proportions during FT context schedules tended to be lower than equivalent FI context schedules (see Fig. 1). However, this was not statistically significant (p = 0.09). Unlike previous reinforcement context studies (Goldshmidt et al., 1998; Williams & Fantino, 1996), we were able to maintain a large range of different response rates across context schedule conditions (Table 1, rightmost column). It may be that initial link choice proportions were affected by some combination of changes in context schedule duration and response rates, regardless of response-reinforcer contingency. If either the change in context duration and/or response rates was not large enough, no significant choice proportion change resulted. For example, response rates were significantly different between the 20 s and 180 s context schedule durations. Correspondingly, initial link response rates between 20 s and 180 s context schedules were significantly different. However, there was only a trend level difference in choice proportions between the baseline and 180 s context schedule durations. In addition to replicating the current findings, future research should target this potential interaction between context schedule response rates and durations by using differential reinforcement of high and low response rate reinforcement schedules.

The moderate effect sizes, combined with individual-level variability suggest that the current experiment may have been underpowered to detect significant effects for each choice proportion and response rate comparison. For example, there was a trend level choice proportion difference between the 20 s and 60 s context schedule conditions. However, this comparison had the same effect size (d= 0.61) as the only significant choice proportion difference between baseline and the 180 s context schedule condition. Future replications should increase sample size to guard against this problem.

As a result of the context schedules' functional similarity to the initial link effect, quantitative choice models could easily be expanded to account for these global contextual choice effects. For example, in delay reduction theory (Fantino, 1969; Squires & Fantino, 1971; Fantino, et al., 1993), T, which represents the average time to primary reinforcement from the onset of the initial links, could be modified to include the average time to primary reinforcement from the onset of the associated context schedule. Thus, T would still include the value of the delay to primary reinforcement in the concurrent chains procedure, in addition to reinforcement context schedule. The challenge would be to mathematically determine the way time to primary reinforcement in the concurrent chains procedure and the context schedule combined. That is, how much weight is given to the context schedule, relative to the reinforcement schedules in the concurrent chains procedure? Similar modifications could be made to the contextual choice (Grace, 1994) and hyperbolic value added models (Mazur, 2001).

Although there was a significant main effect for context schedule duration to affect choice proportions, baseline choice proportions were variable both within and across pigeons (Fig. 1). Additionally, two pigeons were dropped from the analyses based on multiple choice proportions below 0.50. It appeared that both pigeons developed a right-side bias. It us unknown whether other pigeons showed a more transitory side-bias that could have influenced the current results. This lack of experimental control during baseline is a limitation of the current results, and therefore the data should be interpreted cautiously. Future global context studies could attempt to increase experimental control by including a return to baseline condition after each context schedule condition. Only pigeons that are successfully reversed to baseline would be allowed to experience the next context schedule condition. Additionally, only five of the 11 pigeons in the current experiment began the experiment with a baseline condition. Thus, establishing a stable baseline before introducing the context schedule may also help to minimize this variability. The blocking design could be partially responsible for overall variability. Transitions between concurrent chains schedule and context schedules occurred once every four reinforcers, potentially increasing carry-over effects across conditions. Lastly, pigeons did not have extensive experience with either FI or FT reinforcement schedules before this experiment. In some cases this resulted in high response rates during the FT schedule when the pigeon first encountered this response-independent schedule (e.g., pigeon EE 175; FT 20 s condition). This may have contributed to some of the overall variability between and within pigeons.

This study is the first to demonstrate that as context schedule duration increases, choice proportions during a concurrent chains procedure tend to decrease. These preliminary findings are consistent with the initial-link effect (Fantino et al., 1993) and behavioral contrast (Williams, 2002), in that initial link response rates tended to increase with increasing context schedule duration. However, given the overall variability in the data, further research is required to replicate the current results, and to determine how influential the initial link contrast effect is for initial link choice proportions.

DOI 10.1007/s40732-016-0204-1

[mail] Paul Romanowich

paul.romanowich@utsa.edu

Alyssa Cozine

ajcozine@gmaii.com

Daniel L. Worthen

dworthen@csuchico.edu

Compliance with Ethical Standards

Funding None

Conflict of Interest All authors declare they have no competing interests.

Ethical Approval Chico State's Animal Care and Use Committee (ACUC) approved all procedures prior to experimentation. Pigeons were cared for according to the standards described in California State University, Chico's Animal Welfare Policy. All applicable international, national, and institutional guidelines for care and use of animals were followed.

Informed Consent No human participants were enrolled.

Published online: 30 September 2016

References

Fantino, E. (1969). Choice and rate of reinforcement. Journal of the Experimental Analysis of Behavior, 12, 723-730.

Fantino, E., & Davison, M. (1983). Choice: Some quantitative relations. Journal of the Experimental Analysis of Behavior, 40, 1-13.

Fantino, E., Preston, R. A., & Dunn, R. (1993). Delay-reduction: Current status. Journal of the Experimental Analysis of Behavior, 60, 159-169.

Fleshier, M., & Hoffman, H. S. (1962). A progression for generating variable-interval schedules. Journal of the Experimental Analysis of Behavior, 5, 529-530.

Goldshmidt, J. N" Lattal, K. M" & Fantino, E. (1998). Context effects on choice. Journal of the Experimental Analysis of Behavior, 70, 301-320.

Grace, R. C. (1994). A contextual model of concurrent-chains choice. Journal of the Experimental Analysis of Behavior, 61, 113-129.

Grace, R. C., & Bragason, O. (2004). Does the terminal-link effect depend on duration or reinforcement rate? Behavioural Processes, 67, 67-79.

Halliday, M. S" & Boakes, R. A. (1971). Behavioral contrast and response independent reinforcement. Journal of the Experimental Analysis of Behavior, 16, 429-434.

Hursh, S. R., & Fantino, E. (1973). Relative delay of reinforcement and choice. Journal ofthe Experimental Analysis of Behavior, 19,437-450.

LaFiette, M. H., & Fantino, E. (1989). Responding on concurrent-chains schedules in open and closed economies. Journal of the Experimental Analysis of Behavior, 51, 329-342.

MacEwen. D. (1972). The effects of terminal-link fixed-interval and variable-interval schedules on responding under concurrent chained schedules. Journal of the Experimental Analysis of Behavior, 18, 253-261.

Mazur, J. E. (2001). Hyperbolic value addition and general models of animal choice. Psychological Review, 108(1), 96-112.

Mazur, J. E" & Fantino, E. (2014). Choice. In F. K. McSweeney & E. S. Murphy (Eds.), The Wiley Blackwell handbook of operant and classical conditioning (pp. 195-220). Oxford: Wiley-Blackwell.

Reynolds, G. S. (1961). Behavioral contrast. Journal of the Experimental Analysis of Behavior, 4, 57-71.

Shull, R. L., & Pliskoff, S. S. (1967). Changeover delay and concurrent schedules: Some effects on relative performance measures. Journal of the Experimental Analysis of Behavior, 10, 517-527.

Squires, N" & Fantino, E. (1971). A model for choice in simple concurrent and concurrent-chains schedules. Journal of the Experimental Analysis of Behavior, 15, 27-38.

Thompson. D. M., & Corr, P. B. (1974). Behavioral parameters of drug action: Signaled and response-independent reinforcement. Journal of the Experimental Analysis of Behavior, 21, 151-158.

Weisman, R. G., & Ramsden, M. (1973). Discrimination of a response-in-dependent component in a multiple schedule. Journal of the Experimental Analysis of Behavior, 19, 55-64.

Williams, B. A. (2002). Behavioral contrast redux. Animal Learning & Behavior, 30, 1-20.

Williams, B. A., & Fantino, E. (1978). Effects on choice of reinforcement delay and conditioned reinforcement. Journal of the Experimental Analysis of Behavior, 29, 77-86.

Williams, W. A., & Fantino, E. (1996). Response-dependent prochoice effects on foraging related choice. Journal of the Experimental Analysis of Behavior, 65. 619-641.

(1) Including pigeons 17 and EA1752 in the analysis resulted in the following results: For initial link choice proportions a repeated measures ANOVA showed no significant main effects for context schedule duration F(3, 30) = 2.49, p = 0.08 or response contingency F(1, 10) = 0.28, p = 0.61, but a significant interaction between context schedule duration and response contingency F(3, 30) = 3.50, p = 0.04, [[eta].sup.2] = 0.26. For initial link, terminal link, and context schedule response rates, a repeated measure ANOVA showed identical results as when the two pigeons were excluded.

Paul Romanowich [1] (ID) * Alyssa Cozine [2] * Daniel L. Worthen [2]

[1] Department of Psychology, University of Texas at San Antonio, One UTSA Circle, San Antonio, TX 78249-0652, USA

[2] Department of Psychology, California State University, Chico, 400 West First Street, Chico, CA 95929-0234, USA

Caption: Fig. 1 Mean initial link choice proportions during each context schedule condition and baseline across the nine subjects that maintained stable choice proportions above 0.50 for each context schedule condition. Each point represents the mean initial link choice proportion for one pigeon, during one condition. Horizontal dashes represent the mean of all pigeons for that condition

Caption: Fig. 2 Mean initial link response rates during each context schedule condition and baseline across the nine subjects that maintained stable choice proportions above 0.50 for each context schedule condition. Each point represents the mean initial link response rate for one pigeon, during one condition. Horizontal dashes represent the mean of all pigeons for that condition
Table 1 Sequence of experimental conditions, number
of trials to stability criterion, initial link choice
proportions, and mean response rate data for each pigeon

Pigeon   Condition   VI 15   Trials   IL
                     TL               Proportion

EA1776   Baseline    L       20       0.92
         FI 60       R       24       0.73
         FT 180      L       22       0.66
         FI 180      R       20       0.68
         FT 20       L       24       0.66
         FI 20       R       20       0.82
         FT 60       L       20       0.52
         Baseline    R       20       0.70
         Baseline    L       20       0.51
EA1889   FT 180      R       24       0.74
         Baseline    L       20       0.82
         FT 60       R       25       0.51
         FI 180      L       20       0.82
         FT 20       R       23       0.61
         FI 20       L       23       0.86
         FI 60       L       20       0.84
         Baseline    R       20       0.63
EB1057   FI 20       R       20       0.85
         Baseline    L       21       0.85
         FI 180      L       20       0.73
         FT 60       R       22       0.69
         FT 20       L       21       0.82
         FT 180      R       22       0.65
         FI 60       L       25       0.76
         Baseline    R       20       0.62
         Baseline    L       22       0.81
4991     Baseline    R       20       0.96
         FT 180      L       23       0.58
         FI 20       R       23       0.87
         FT 20       L       20       0.70
         FI 180      R       21       0.78
         FI 60       L       22       0.69
         Baseline    L       20       0.66
         FT 60       R       21       0.70
EA1857   FI 20       L       20       1.00
         FT 20       R       21       0.65
         Baseline    L       20       0.97
         FT 180      R       22       0.56
         FI 60       R       20       0.64
         FI 180      L       24       0.77
         FT 60       L       24       0.89
         Baseline    R       20       0.63
         Baseline    L       20       0.78
DY1006   Baseline    R       20       0.99
         FI 20       L       20       0.72
         FI 60       R       24       0.87
         FT 60       L       21       0.73
         FI 180      R       20       0.80
         Baseline    L       20       0.61
         FT 20       R       20       0.74
         FT 180      L       27       0.72
EE 175   FT 20       L       20       0.84
         Baseline    R       35       0.84
         FI 60       L       20       0.66
         FI 20       R       20       0.71
         FT 180      L       20       0.71
         FT 60       R       27       0.72
         FI 180      R       20       0.57
         Baseline    L       20       0.74
EA1752   FT 60       L       20       0.92
         Baseline    R       20       0.80
         FI 180      L       25       0.45
         FI 20       R       20       0.77
         FI 60       R       20       0.83
         FT 20       L       20       0.46
         FT 180      R       20       0.70
         Baseline    L       20       0.52
         Baseline    R       21       0.66
EB1066   FT 20       R       20       0.79
         FT 60       L       23       0.67
         Baseline    R       22       0.63
         FT 180      L       21       0.80
         FI 60       R       29       0.67
         FI 20       L       23       0.75
         FI 180      R       26       0.60
         Baseline    L       23       0.87
EA1731   Baseline    R       20       0.81
         FI 20       L       20       0.58
         FI 60       R       23       0.67
         FT 180      L       24       0.64
         FI 180      R       28       0.74
         FT 60       L       22       0.66
         Baseline    L       20       0.73
         FT 20       R       23       0.77
17       Baseline    L       20       0.77
         FT 180      R       20       0.69
         FI 20       L       23       0.47
         FI 180      R       20       0.74
         FT 20       R       22       0.79
         FT 60       L       20       0.35
         FI 60       R       22       0.80
         Baseline    R       20       0.73
         Baseline    L       22       0.45

Pigeon   Condition   Mean response rate (resp./min)

                     Left      Right    VI 15     VI45      Context
                     IL        IL       TL        TL

EA1776   Baseline     55.72     4.72    124.72     84.32
         FI 60        27.02    76.06    140.70     99.64      67.80
         FT 180       67.24    35.06    154.82    119.14       0.00
         FI 180       40.94    87.24    138.08    118.38      56.00
         FT 20        55.08    27.86    120.40    129.74       0.00
         FI 20        20.00    90.24    159.20     87.46      86.60
         FT 60        64.58    59.08    129.62    137.64       0.60
         Baseline     27.08    62.64    121.84    120.22
         Baseline     50.70    48.32    156.46    110.68
EA1889   FT 180        9.08    26.08     74.00     57.10       0.00
         Baseline     37.06     7.94     92.50     56.26
         FT 60        25.30    26.26     57.98     79.98       0.00
         FI 180       40.36     9.10     97.04     54.32      44.10
         FT 20        16.02    24.56    104.02     73.04       0.00
         FI 20        28.08     4.66     95.94     66.58      98.14
         FI 60        36.80     7.06    108.06     82.66      57.86
         Baseline     14.90    24.94     87.40     77.40
EB1057   FI 20        12.90    73.48     70.18     98.50      42.78
         Baseline     81.72    14.98     92.68     47.22
         FI 180       79.10    29.40    106.78     35.30      57.38
         FT 60        26.62    59.60     47.20    124.06       9.24
         FT 20        80.98    17.90    102.88     54.40       1.28
         FT 180       32.76    61.66     74.42     81.92       7.86
         FI 60        63.12    19.50     79.68     67.30      49.68
         Baseline     30.26    49.24     68.18     61.24
         Baseline     57.92    13.58     73.98     65.20
4991     Baseline      2.04    53.38     64.44     90.44
         FT 180       39.60    28.22    103.50     63.90      17.78
         FI 20         1.88    12.78     36.84     23.30      19.68
         FT 20        46.60    19.86     99.90     74.34      30.18
         FI 180       23.22    82.24     92.36     90.88      35.70
         FI 60        54.55    24.70    107.88     82.00      47.83
         Baseline     55.56    28.96    107.76     70.98
         FT 60        26.94    62.34     96.76     14.74      20.72
EA1857   FI 20        55.48     0.02    103.82      0.00      60.94
         FT 20        19.44    35.92    103.22     76.34      49.06
         Baseline     64.48     2.48     93.98     77.88
         FT 180       31.76    39.88    112.56     86.38       1.22
         FI 60        25.02    44.40    112.86     82.98      68.04
         FI 180       52.48    15.98    103.04     87.76      49.96
         FT 60        54.02     6.82     93.76     86.50      21.80
         Baseline     23.82    40.90     97.64     86.84
         Baseline     46.30    12.96     92.36     79.02
DY1006   Baseline      0.40    44.40    112.08     79.02
         FI 20        43.20    16.78    115.82     75.06     118.18
         FI 60        10.24    66.34    125.32     95.46      65.00
         FT 60        58.04    21.18    111.28     99.80       0.26
         FI 180       16.48    67.60    124.26     93.74      45.30
         Baseline     45.90    29.38    120.14     84.66
         FT 20        18.22    51.86    148.62     72.10       0.00
         FT 180       52.74    20.60    122.94     90.68       0.32
EE 175   FT 20        48.16     9.36    106.58    113.50     128.28
         Baseline     15.86    84.65    177.40     88.64
         FI 60        61.02    32.08     84.72    125.72     104.22
         FI 20        22.42    54.92    129.56     76.72      37.54
         FT 180       68.26    27.62     83.38    115.44       0.12
         FT 60        23.42    60.10    138.02     67.52       0.96
         FI 180       26.94    35.80    149.66     65.08      38.18
         Baseline     51.84    18.68     62.96     78.62
EA1752   FT 60        45.80     3.88    105.18     60.40       4.90
         Baseline     13.42    54.38     91.08     66.46
         FI 180       38.78    46.88     93.26     74.86      36.24
         FI 20        21.14    70.06     91.80     69.98      47.32
         FI 60        17.80    87.68    100.70     76.80      30.72
         FT 20        49.50    57.88     93.08     71.38      45.78
         FT 180       35.12    79.90     82.58     74.96       2.62
         Baseline     53.44    49.92     87.34     61.04
         Baseline     38.94    75.94     78.16     68.64
EB1066   FT 20         8.52    32.58    194.24    125.44       0.80
         FT 60        52.86    25.56    175.68    109.48       2.14
         Baseline     37.36    63.94    166.10    122.20
         FT 180      107.30    26.48    206.72    106.88       3.66
         FI 60        44.06    91.42    178.64    132.60      65.10
         FI 20       101.24    33.16    180.80     76.94      98.96
         FI 180       54.60    81.90    130.14    137.00      45.22
         Baseline    100.36    15.48    186.50    100.22
EA1731   Baseline     10.68    46.22    124.14     65.90
         FI 20        23.28    16.58    132.86     66.92      61.70
         FI 60        22.88    47.50    117.60     61.62      47.04
         FT 180       53.34    29.52    109.08     80.18       1.00
         FI 180       22.22    62.50    112.78     79.86      34.70
         FT 60        57.10    29.86    109.54     72.06       0.04
         Baseline     54.06    20.32    114.58     68.68
         FT 20        17.58    59.86     96.92     61.88       1.02
17       Baseline     65.70    19.44    169.58    138.88
         FT 180       23.62    52.24    139.56    146.48       0.26
         FI 20        35.72    40.06    144.08    129.38      91.56
         FI 180       18.88    55.08    123.24    128.24      68.60
         FT 20        14.56    54.16    145.30    116.62     145.10
         FT 60        21.30    39.60    126.44    121.58      10.02
         FI 60        11.30    45.38    140.32    108.24      52.64
         Baseline     14.06    38.04    118.40     84.34
         Baseline     20.30    25.03     82.43     89.40
COPYRIGHT 2017 Springer
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2017 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Title Annotation:ORIGINAL ARTICLE
Author:Romanowich, Paul; Cozine, Alyssa; Worthen, Daniel L.
Publication:The Psychological Record
Article Type:Report
Date:Mar 1, 2017
Words:6022
Previous Article:Categorical discrimination of sequential stimuli: all [S.sup.[DELTA]] are not created equal.
Next Article:Laboratory measures of aggression in methadone patients pre- and postdose.
Topics:

Terms of use | Privacy policy | Copyright © 2019 Farlex, Inc. | Feedback | For webmasters