Human temporal learning with mixed signals.

The influence of cue informativeness on human temporal discrimination was evaluated using a peak-interval (PI) procedure. A target moved across the computer monitor, reaching the center at 2 or 4 s. Key presses shot the center of the screen. Participants earned points when shots hit the target and lost points for misses. The target was masked during occasional, extended PI trials, allowing for measurement of temporal discrimination. During PI trials, the screen background color could exert stimulus control by providing information about target speed. Cue informativeness was represented as the correlation (φ) between light or dark green backgrounds and the 2- or 4-s target and was manipulated across 4 conditions: a multiple schedule (φ = 1), mixed signals (φ = 0.8, 0.4), and a mixed schedule (φ = 0). In Experiment 1, participants were randomly assigned to one of the 4 conditions. In Experiment 2, each participant experienced all 4 conditions. Participants learned to respond at both intervals in all conditions. Cue informativeness did not affect peak time or spread. For the most part, temporal distributions of responses for the two background colors suggested a cover-both-bases strategy in the presence of mixed signals. Participants incorporated probabilistic information from cues to allocate responding in time.

Beyond inference by eye: Statistical and graphing practices in JEAB, 1992-2017.

Debates about the utility of p values and correct ways to analyze data have inspired new guidelines on statistical inference by the American Psychological Association (APA) and changes in the way results are reported in other scientific journals, but their impact on the Journal of the Experimental Analysis of Behavior (JEAB) has not previously been evaluated. A content analysis of empirical articles published in JEAB between 1992 and 2017 investigated whether statistical and graphing practices changed during that time period. The likelihood that a JEAB article reported a null hypothesis significance test, included a confidence interval, or depicted at least one figure with error bars has increased over time. Features of graphs in JEAB, including the proportion depicting single-subject data, have not changed systematically during the same period. Statistics and graphing trends in JEAB largely paralleled those in mainstream psychology journals, but there was no evidence that changes to APA style had any direct impact on JEAB. In the future, the onus will continue to be on authors, reviewers and editors to ensure that statistical and graphing practices in JEAB continue to evolve without interfering with characteristics that set the journal apart from other scientific journals.

Selective attention in pigeon temporal discrimination.

Cues can vary in how informative they are about when specific outcomes, such as food availability, will occur. This study was an experimental investigation of the functional relation between cue informativeness and temporal discrimination in a peak-interval (PI) procedure. Each session consisted of fixed-interval (FI) 2- and 4-s schedules of food and occasional, 12-s PI trials during which pecks had no programmed consequences. Across conditions, the phi (ϕ) correlation between key light color and FI schedule value was manipulated. Red and green key lights signaled the onset of either or both FI schedules. Different colors were either predictive (ϕ = 1), moderately predictive (ϕ = 0.2-0.8) or not predictive (ϕ = 0) of a specific FI schedule. This study tested the hypothesis that temporal discrimination is a function of the momentary conditional probability of food; that is, pigeons peck the most at either 2 s or 4 s when ϕ = 1 and peck at both intervals when ϕ < 1. Response distributions were bimodal Gaussian curves; distributions from red- and green-key PI trials converged when ϕ ≤ 0.6. Peak times estimated by summed Gaussian functions, averaged across conditions and pigeons, were 1.85 and 3.87 s; however, pigeons did not always maximize the momentary probability of food. When key light color was highly correlated with FI schedules (ϕ ≥ 0.6), estimates of peak times indicated that temporal discrimination accuracy was reduced at the unlikely interval, but not the likely interval. The mechanism of this reduced temporal discrimination accuracy could be interpreted as an attentional process.

Tutorial: Small-N Power Analysis.

Power analysis is an overlooked and underreported aspect of study design. A priori power analysis involves estimating the sample size required for a study based on predetermined maximum tolerable Type I and II error rates and the minimum effect size that would be clinically, practically, or theoretically meaningful. Power is more often discussed within the context of large-N group designs, but power analyses can be used in small-N research and within-subjects designs to maximize the probative value of the research. In this tutorial, case studies illustrate how power analysis can be used by behavior analysts to compare two independent groups, behavior in baseline and intervention conditions, and response characteristics across multiple within-subject treatments. After reading this tutorial, the reader will be able to estimate just noticeable differences using means and standard deviations, convert them to standardized effect sizes, and use G*Power to determine the sample size needed to detect an effect with desired power.

Probability discounting and gambling: a meta-analysis.

BACKGROUND AND AIMS: A number of studies have investigated connections between probability discounting and gambling. The aim of this research was to obtain a meta-analytical weighted effect size for the relationship between shallow probability discounting (the tendency to overvalue reinforcement with lower odds) and gambling intensity and to examine whether a gambling diagnosis moderated this effect size such that the relationship is stronger for diagnosed problem gamblers. METHODS: A database search identified studies that (a) measured both probability discounting and gambling and (b) reported statistical results allowing calculation of an effect size for meta-analysis. The search resulted in 12 studies reporting statistical results for probability discounting and gambling. The studies comprised 1685 individuals from different cohorts and nations, and included gamblers and non-gamblers. The studies reported 18 effect sizes. Across studies, gambling severity was assessed through diagnosis and gambling intensity was assessed through self-report and performance. Comprehensive Meta Analysis software calculated the weighted effect size and moderating role of gambling diagnosis. RESULTS: Shallower probability discounting was associated with greater gambling severity or intensity in all 12 studies. Throughout the studies, the weighted meta-analytical effect size for the connection between probability discounting and gambling was significant, with Hedges' g = 0.36 [standard error (SE) = 0.07, 95% confidence interval (CI) = 0.21, 0.50), P < 0.001]. Addressing the second aim of the study, individuals diagnosed with a gambling disorder or problem gambling compared with not diagnosed individuals showed an effect size of Hedges' g = 0.79 (SE = 0.18, 95% CI = 0.45, 1.14) and a moderation analysis indicated that this type of comparison showed significantly stronger effects than effect sizes based on associations between probability discounting and gambling (Q(1) = 7.80, P = 0.005). CONCLUSIONS: There appears to be a positive association between problem gambling and shallow probability discounting (a cognitive bias that overvalues low probability gains and/or undervalues high probability losses).

Differential reinforcement of low rates differentially decreased timing precision.

Timing processes have been implicated as potential mechanisms that underlie self-controlled choice. To investigate the impact of an intervention that has been shown to increase self-controlled choice on timing processes, accuracy and precision of temporal discrimination were assessed in an 18-s peak procedure (18-s fixed interval trials; 54-s peak trials). During an intervention phase, mice in three treatment groups experienced differential reinforcement of low rate (DRL) schedules of reinforcement of 27 s, 18 s, or 9 s. A fourth group received continued exposure to the peak procedure. After the DRL intervention, timing was reassessed using the peak procedure. In contrast to previous reports, the DRL intervention resulted in less precise timing as indicated by increased peak spread and disrupted single-trial measures of temporal control. These effects were only detected just after the DRL intervention suggesting a transient effect of DRL exposure on timing. The increase in peak spread in the present experiment suggests delay exposure via DRL schedules may produce a "dose-dependent" effect on temporal discrimination, which may also increase self-controlled choice.

Adaptive learning and forgetting in an unconventional experimental routine.

Forgetting is often thought of as the inability to remember, but remembering and forgetting allow behavior to adapt to a changing environment in distinct and separable ways. Learning and forgetting were assessed concurrently in two pigeon experiments that involved the same unconventional routine where the schedule of reinforcement changed every session. Sessions were run back-to-back with a 23-h mid-session break such that in a single visit to the testing chamber, a pigeon completed the second half of one session and the first half of the next. The beginning of a new session was either signaled or unsignaled. Experiment 1 involved concurrent variable-interval variable-interval schedules with four possible reinforcer ratios. Response allocation was sensitive to the richer schedule and was retained through the mid-session break. Experiment 2 involved peak interval schedules of varying durations. Temporal discrimination was rapidly acquired before and after the mid-session break, but not retained. Signaling the session change decreased control by past contingencies in both experiments, demonstrating that learning and forgetting can be investigated separately. These results suggest that the temporal structure of training, such as multiple short daily sessions instead of one long session, can meaningfully impact measurement of animals' capacity to forget and remember.

Translating Behavior Analysis: a Spectrum Rather than a Road Map.

Much has been written about the potential benefits of translational research in behavior analysis, but a lack of consensus about what constitutes "translational" creates a barrier to effective knowledge-practice innovation within the discipline and across other sciences. We propose a tiered system, adapted from a biomedical translational pathway, for classifying behavior analysis research on a basic-applied spectrum. Tier 0 is blue sky basic science in which the subjects, behaviors, stimuli, and settings are selected for convenience. Tier 1 is use-inspired basic science with a socially important end game and research subject. Tier 2 is solution-oriented research that attempts to solve a specific problem in a socially important subject, but 1 or more aspects of the research are selected for purposes of experimental control rather than social importance. Tier 3 is applied behavior analysis research that studies a problem of social significance for the subject and involves behaviors, stimuli, and settings that are socially important. Tier 4 is impact assessment in which behavioral technology is applied with a direct benefit to society. We provide examples of behavior-analytic research in each tier and evaluate the potential benefits of organizing behavior analysis in this way.

Searching for the variables that control human rule-governed "insensitivity".

Verbal rules or instructions often exert obvious and meaningful control over human behavior. Sometimes instructions benefit the individual by enabling faster acquisition of a skill or by obviating an aversive consequence. However, research has also suggested a clear disadvantage: "insensitivity" to changing underlying contingencies. The two experiments described here investigated the variables that control initial rule-following behavior and rule-following insensitivity. When the initial rule was inaccurate, behavior was consistent with the rule for approximately half of participants and all participants' behavior was mostly insensitive to changing contingencies. When the initial rule was accurate, behavior of all participants was consistent with it and behavior for nearly all participants was insensitive to changes in underlying contingencies. These findings have implications for how best to establish and maintain rule-following behavior in applied settings when deviant behavior would be more reinforcing to the individual.

Effects of signaling on temporal control of behavior in response-initiated fixed intervals.

Behavior and events distributed in time can serve as markers that signal delays to future events. The majority of timing research has focused on how behavior changes as the time to some event, usually food availability, decreases. The primary objective of the two experiments presented here was to assess how behavior changes as time passes between two time markers when the first time marker was manipulated but the second, food delivery, was held constant. Pigeons were exposed to fixed-interval, response-initiated fixed-interval, and signaled response-initiated fixed-interval 15- and 30-s schedules of reinforcement. In Experiment 1, first-response latencies were systematically shorter in the signaled response-initiated schedules than response-initiated schedules, suggesting that the first response was a more effective time marker when it was signaled. In Experiment 2, responding in no-food (i.e. "peak") trials indicated that timing accuracy was equivalent in the three schedule types. Compared to fixed interval schedules, timing precision was reduced in the signaled response-initiated schedules and was lowest in response-initiated schedules. Results from Experiments 1 and 2 coupled with previous research suggest that the overall "informativeness" of a time marker relative to other events and behaviors in the environment may determine its efficacy.

What is timed in a fixed-interval temporal bisection procedure?

Recent research on interval timing in the behavioral and neurological sciences has employed a concurrent fixed-interval (FI) procedure first reported by Platt and Davis (Journal of Experimental Psychology: Animal Behavior Processes, 9, 160-170, 1983). Studies employing the task typically assess just 1 dependent variable, the switch/bisection point; however, multiple measures of timing are available in the procedure and it is unclear (a) what is timed (i.e., learned) by subjects and (b) what other measures might tell us about timing in the task and generally. The main objective of the current experiment was to utilize multiple dependent measures of timing accuracy and precision derived from the task to assess whether the 2 FIs are timed independently or if timing 1 FI interferes with timing the other, and vice versa. Four pigeons were exposed to an FI temporal bisection procedure with parametric manipulations across two phases. In the constant phase, the short FI was always the same; the long FI was 2 to 16 times the short FI and changed across conditions. In the proportional phase, the long FI was always 4 times the duration of the short FI. Across both phases, pigeon mean bisection points were near the geometric mean of the 2 FIs. Coefficients of variance increased as the durations to be timed increased. Results suggested pigeons' timing of the short FI was affected by the presence of the long FI, and vice versa. The FI temporal bisection task offers multiple dependent variables for analysis and is well suited for studying temporal learning and decision making.

Environmental dynamics modulate covariation of choice and timing.

Response allocation between delayed reinforcers is presumably a function of the discrimination of those delays. In the present experiment, we analyzed the functional relation between response allocation and temporal discrimination across different environmental dynamics. Three pigeons pecked for food in a concurrent-chain schedule. Concurrent variable-interval initial links produced fixed-interval (FI) terminal links. Start and stop times, single-trial measures of temporal discrimination, were obtained from occasional 'no-food' terminal links. In dynamic, rapid-acquisition conditions, terminal links were FI 10s and 20s and the location of the initial link leading to the shorter terminal link varied unpredictably across sessions. In the static conditions, both terminal links were either "uniform" FI 15-s schedules or one terminal link was "fixed" at FI 10s and the other at 20s. Response allocation and start and stop times adjusted within sessions in dynamic conditions and across sessions of static conditions. Residuals from regressions of expected on programmed immediacy ratios were positively correlated to a greater magnitude in dynamic than static conditions. This change in residual covariation demonstrated that environmental dynamics modulated the relation between choice and timing.

Timing in response-initiated fixed intervals.

Different events can serve as time markers that initiate intervals in schedules of reinforcement. Pigeons were exposed to fixed-interval (FI) schedules in which the onset of the interval was signaled by the illumination of a key light or initiated by a peck to a lighted key. Food was delivered following the first response after the interval elapsed. In Experiment 1, three pigeons were exposed to a multiple schedule. One component was a standard FI schedule: Key light illumination signaled the onset of the interval. The other component was a response-initiated fixed-interval (RIFI) schedule: The first key-peck response determined the onset of the interval. In Experiment 2, three pigeons were exposed to a multiple FI-RIFI schedule of reinforcement and on occasional trials food was not delivered (i.e. "no-food" or "peak trials"). A yoking procedure equated reinforcement rates between the schedule types in both experiments. Absolute response rates early in the intervals were higher in the RIFI schedules of both experiments. Normalized response-rate gradients, ogive fits of normalized response gradients, and breakpoints were not systematically different for the schedule types in Experiment 1, indicating similar patterns of responding between interval onset and food delivery. However, during peak trials in Experiment 2 the duration of responding at a high rate was longer for RIFI schedules than FI schedules. This suggests that timing precision was reduced in the RIFI schedules and that relative "distinctiveness" of a time marker may determine its efficacy.

Choice and timing in pigeons under differing levels of food deprivation.

State-dependent valuation learning (SDVL) is a preference for stimuli associated with relative food deprivation over stimuli associated with relative satiety. Pigeons were exposed to experimental conditions designed to investigate SDVL and to test the hypothesis that obtained relative immediacy during training predicts choice during test probes. Energy states were manipulated using a procedure that has previously revealed SDVL in starlings and pigeons. In Experiment 1, pigeons preferred the stimulus associated with deprivation in the first choice probe session, but were indifferent in the second. Changes in choice were consistent with changes in obtained relative immediacy. In Experiment 2, training parameters were altered and SDVL did not occur. Obtained relative immediacy again predicted choice. Results of both experiments provide evidence that obtained relative immediacy may be an important contributing factor to the SDVL phenomenon.

Quantifying transitions in response allocation with change point analysis in concurrent chains.

Research based on the matching law has demonstrated empirically that the physical and temporal properties of the events, the context in which they occur and the signals that mark them in space and time all contribute to response allocation. When the physical or temporal properties of different outcomes change in ways that affect their relative value, the ratio of responses to each outcome adjusts with time and exposure to the new contingency. Five pigeons pecked in concurrent-chain schedules with fixed-interval terminal links. Terminal-link schedules were changed each session. In most sessions, response allocation was initially indifferent to terminal-link schedules but shifted to favor the initial link associated with the shorter terminal link. As a first step to disambiguating response allocation in transition from stable response allocation, transition durations were interpolated from change points in cumulative response plots. The relation between transition duration and absolute log immediacy ratio was negative: the number of initial links until the shift occurred was longer when terminal-link schedules were relatively similar than when they were relatively different.

Pigeons learn signal-food intervals independently in a multiple peak procedure.

Previous research has shown rapid learning of multiple temporal relations between signals and food by pigeons when these relations are changed unpredictably each session (Kyonka & Grace, 2007). The goal of the present study was to test whether contextual temporal cues-that is, an alternative signal-food delay that was a valid predictor of a target signal-food delay-facilitated acquisition by the target contingency. Four pigeons responded in a multiple peak-interval procedure in which red and green keys signaled separate fixed-interval (FI) schedules with occasional extinction probes (peak trials). The schedule parameters of the FIs either summed to 30 s (correlated condition; ρ = -1.0) or were not restricted to sum to 30 s (uncorrelated condition; ρ = 0.0). Comparing stop times obtained from peak trials in the 2 conditions revealed no effect of context: Temporal control of responding was acquired at the same rate and with the same precision regardless of whether the schedule values were correlated. These results suggest that pigeons learn about multiple signal-food delays independently.

Pigeon responding in fixed-interval and response-initiated fixed-interval schedules.

In fixed-interval (FI) and response-initiated fixed-interval (RIFI) schedules of reinforcement, a response is required after an interval has elapsed for delivery of reinforcement. In RIFI schedules, a response is required to initiate each interval as well. The objective of this experiment was a systematic comparison of performance in the two schedule types over a range of interval durations. Four pigeons were exposed to FI and RIFI schedules of 15, 30, 60, 120 and 240 s. Interfood intervals were longer and more variable in RIFI than corresponding FI schedules. In addition, response rates early in the RIFI schedules were higher than in corresponding FI schedules. However, the distribution of first-response latencies, mean breakpoints, and normalized response gradients suggest that temporal discrimination was equivalent in the two schedules.

Rapid acquisition of choice and timing and the provenance of the terminal-link effect.

Eight pigeons responded in a concurrent-chains procedure in which terminal-link schedules changed pseudorandomly across sessions. Pairs of terminal-link delays either summed to 15 s or to 45 s. Across sessions, the location of the shorter terminal link changed according to a pseudorandom binary sequence. On some terminal links, food was withheld to obtain start and stop times, measures of temporal control. Log initial-link response ratios stabilized within the first half of each session. Log response ratio was a monotonically-increasing but nonlinear function of programmed log terminal-link immediacy ratio. There was an effect of absolute terminal-link duration on log response ratio: For most subjects, preference for the relatively shorter terminal-link delay was stronger when absolute delays were long than when absolute delays were short. Polynomial regressions and model comparison showed that differences in degree of nonlinearity, not in sensitivity to log immediacy ratio, produced this effect. Temporal control of stop times was timescale invariant with scalar variability, but temporal control of start times was not consistent across subjects or terminal-link durations.

Effects of unpredictable changes in initial-link duration on choice and timing.

Four pigeons responded in a concurrent-chains procedure in which terminal-link schedules were fixed-interval (FI) 10s and FI 20s. Across sessions, the location of the shorter terminal-link changed according to a pseudorandom binary sequence. Each session, the variable-interval initial-link schedule value was sampled from a uniform distribution that ranged from 0.01 to 30s. On some terminal links, food was withheld to obtain measures of temporal control. Terminal-link delays determined choice (log initial-link response ratios) and timing (start and stop times on no-food trials) measures, which stabilized within the 1st half of each session. Preference for the shorter terminal-link delay was a monotonically decreasing function of initial-link duration. There was no evidence of control by initial-link durations from previous sessions.