ASSESSING OBSERVER ACCURACY IN CONTINUOUS RECORDING OF RATE AND DURATION: THREE ALGORITHMS COMPARED

The three algorithms most frequently selected by behavior‐analytic researchers to compute interobserver agreement with continuous recording were used to assess the accuracy of data recorded from video samples on handheld computers by 12 observers. Rate and duration of responding were recorded for three samples each. Data files were compared with criterion records to determine observer accuracy. Block‐by‐block and exact agreement algorithms were susceptible to inflated agreement and accuracy estimates at lower rates and durations. The exact agreement method appeared to be overly stringent for recording responding at higher rates (23.5 responses per minute) and for higher relative duration (72% of session). Time‐window analysis appeared to inflate accuracy assessment at relatively high but not at low response rate and duration (4.8 responses per minute and 8% of session, respectively).     

USING CALIBRATION AND INTEROBSERVER AGREEMENT ALGORITHMS TO ASSESS THE ACCURACY AND PRECISION OF DATA FROM ELECTRONIC AND PEN‐AND‐PAPER CONTINUOUS RECORDING METHODS

Often it is assumed that electronic recording by observers necessarily provides better quality data than pen‐and‐paper methods. Fifteen novice observers recorded rates of responding from 10 role‐played video samples using one of three continuous recording input formats: keyboard (laptop), touchscreen (personal digital assistants), or pen‐and‐paper. We evaluated the quality of the observers' data compared with criterion records using calibration and interobserver agreement algorithms. Results of the calibration analysis revealed that observers in the touchscreen group produced the most consistently accurate and precise data, the keyboard group observers showed wide variation in precision and accuracy, and the pen‐and‐paper group observers were significantly less precise than the touchscreen group. We conclude that although electronic recording has the potential to be as accurate as, and more precise than, pen‐and‐paper methods, this is far from guaranteed. Analyses of observers' errors advise recommendations for improving data accuracy and precision when using each method. Copyright © 2014 John Wiley & Sons, Ltd.     

Rates of responding measured by continuous recording in applied behavioral research

There is no published source that can inform behavioral measurement research about typical response rates measured by continuous recording in applied behavior analysis. The present study was designed to determine the distribution of data on rate of responding reported across a recent 10‐volume (1998–2007) sample of research articles from the Journal of Applied Behavior Analysis. Rate of responding was measured for 20 553 datapoints in 360 sets of data (one participant in one graph) presented in 60 research articles. The majority of datapoints showed response rates of less than 1.0 response per minute, with 28.3% at zero. The highest rate reported was 104.5 responses per minute. The maximum response rate in datasets was more than 1.0 response per minute in 96.7% of datasets, indicating that continuous recording has been employed in research in which a relatively high rate of responding was recorded at some time during the study. The data from this report will assist researchers to determine empirically the representativeness of response rates in studies of measurement in applied behavior analysis. Copyright © 2010 John Wiley & Sons, Ltd.     

Shaping data collection congruent with experimental hypotheses

The influence of instructions and feedback from an experimenter on observational recordings of disruptive behavior was evaluated. Four subject-observers recorded four categories of disruptive behavior from videotapes of children in a classroom setting. Two sets of videotapes, labelled "baseline" and "treatment", were matched for rates of disruptive behavior in each category. The observers were told that two target behaviors were expected to decrease and the other two control behaviors were not expected to change during the treatment condition. During observational recording of treatment videotapes, the observers were given positive feedback when they reported decreases in the two target behaviors, and negative feedback when they reported either to change or increases in the two control behaviors. The target behaviors were recorded less frequently by observers, while recordings of control behaviors were unaffected during the treatment phase. These results suggest that contingent feedback to observers should be avoided as a possible source of bias in behavioral recordings.     

Decreasing classroom misbehavior through the use of DRL schedules of reinforcement

In three studies, reinforcing low rates of responding reduced inappropriate behaviors. In the first study, the talking-out behavior of one TMR student was reduced when the teacher allowed 5 min of free time for a talk-out rate less than 0.06 per minute. In a second study, the talking-out behavior of an entire TMR class was reduced when reinforcement was delivered for a response rate less than 0.10 per minute. In a third study, successively decreasing DRL limits were used to reduce off-task verbalizations of an entire high school business class. In each case, the DRL procedure proved manageable for the teacher and successful in reducing misbehavior.     

INFLUENCES OF RESPONSE RATE AND DISTRIBUTION ON THE CALCULATION OF INTEROBSERVER RELIABILITY SCORES

We examined the effects of several variations in response rate on the calculation of total, interval, exact‐agreement, and proportional reliability indices. Trained observers recorded computer‐generated data that appeared on a computer screen. In Study 1, target responses occurred at low, moderate, and high rates during separate sessions so that reliability results based on the four calculations could be compared across a range of values. Total reliability was uniformly high, interval reliability was spuriously high for high‐rate responding, proportional reliability was somewhat lower for high‐rate responding, and exact‐agreement reliability was the lowest of the measures, especially for high‐rate responding. In Study 2, we examined the separate effects of response rate per se, bursting, and end‐of‐interval responding. Response rate and bursting had little effect on reliability scores; however, the distribution of some responses at the end of intervals decreased interval reliability somewhat, proportional reliability noticeably, and exact‐agreement reliability markedly.     

Operant and nonoperant vocal responding in the mynah: Complex schedule control and deprivation-induced responding

Several recent studies have been concerned with operant responses that are also affected by nonoperant factors, (e.g., biological constraints, innate behavior patterns, respondent processes). The major reason for studying mynah vocal responding concerned the special relation of avian vocalizations to nonoperant emotional and reflexive systems. The research strategy was to evaluate operant and nonoperant control by comparing the schedule control obtained with the vocal response to that characteristic of the motor responses of other animals. We selected single, multiple, and chain schedules that ordinarily produce disparate response rates at predictable times. In multiple schedules with one component where vocal responding (“Awk”) was reinforced with food (fixed-ratio or fixed-interval schedule) and one where the absence of vocal responding was reinforced (differential reinforcement of other behavior), response rates never exceeded 15 responses per minute, but clear schedule differences developed in response rate and pause time. Nonoperant vocal responding was evident when responding endured across 50 extinction sessions at 25% to 40% of the rate during reinforcement. The “enduring extinction responding” was largely deprivation induced, because the operant-level of naive mynahs under food deprivation was comparable in magnitude, but without deprivation the operant level was much lower. Food deprivation can induce vocal responding, but the relatively precise schedule control indicated that operant contingencies predominate when they are introduced.     

Drug effects in squirrel monkeys trained on a multiple schedule with a punishment contingency

The behavior of four monkeys trained on a multiple schedule was differentially sensitive to selected pharmacological agents. The three components of the multiple schedule were: (1) a variable-interval schedule in which responses were reinforced on the average of once per minute; (2) a concurrent schedule in which every tenth response was reinforced and every fifteenth response, on the average, was shocked; and, (3) a neutral stimulus in the presence of which responses were neither reinforced nor shocked. Pentobarbital, chlordiazepoxide, and meprobamate increased responding during each of the components. Scopolamine and d-amphetamine decreased variable-interval performance, had minimal effects on performance during the concurrent-schedule component, and increased responding in the presence of the neutral stimulus. Chlorpromazine decreased variable-interval responding and had slight effects on the responding during the other two components.     

Factors influencing responding under multiple schedules of conditioned and unconditioned reinforcement

Two experiments examined pigeons' responses under multiple schedules of conditioned and unconditioned reinforcement. In one component, responses produced food according to a fixed-interval schedule; in a second component, responses produced brief stimuli according to a fixed-ratio schedule. When brief-stimulus presentations were paired with food in the first component, rates in the second component were usually higher than 10 responses per minute. When pairing in the first component was eliminated, responding continued to be maintained in the second component. Elimination of food presentation from the first component substantially decreased responding in the second component, even though the brief stimulus had not been paired with food. Experiment II demonstrated that response rate was affected by the duration of both the second component and the brief stimulus. The results suggest that three conditions are important in maintaining responding with brief-stimulus presentations: (1) pairing the brief stimulus, at least initially, with food, (2) maintaining unconditioned reinforcement in one component, and (3) employing optimal brief-stimulus and component durations.     

The measurement of response force during a lever-press shock-escape procedure in rats

Three albino rats were given extensive exposure to an escape procedure in which shocks were scheduled to occur 30 sec apart. The amount of downward force exerted on the lever was continuously recorded and compared with traditional discrete measurement in which only responses above an arbitrary force threshold were recorded. Subjects typically remained in contact with the lever throughout the shock-free intertrial interval. Shock onset reliably occasioned a brief lurch from and return to the lever, which resulted in a lever-press escape response being recorded. Lever contact, which occupied an average of 90% of session time, showed great stability in force after a brief (e.g., 5-sec) period of instability after each shock terminated. In general, continuous measurement revealed considerably different results than discrete measurements of on-lever behavior. For example, continuous lever contact fluctuating near response threshold was often recorded as discrete responding, and sustained lever contact occurring below response threshold was omitted from discrete measurement.     

Drug effects on fixed-interval responding with pause requirements for food presentation.

In pigeons performing under a multiple schedule of food presentation, low key-pecking rates (0.18 to 0.29 responses per second) were maintained during 3-min fixed-interval components by requiring a 4-, 5-, or 6-sec pause preceding the food-delivery response (tandem DRL), while higher rates (0.70 to 1.37 responses per second) were maintained in alternative fixed-interval components by requiring a pause of no more than 40 msec preceding the food-delivery response (tandem DRH). Thus, reinforcement density was equal but overall response rates markedly different in the two schedule components. Pentobarbital (3, 10 mg/kg) had effects on overall rates of responding consistent with a rate-dependency interpretation (low rates were increased while higher rates were decreased), but d-amphetamine (0.03 to 3 mg/kg) either failed to increase low overall rates in the tandem DRL components or increased them only slightly. Effects of both drugs on local responding within the fixed-intervals were always related in an orderly way to control response rate, but the extent of rate increases produced by d-amphetamine was modifed in some birds by pause requirements such that the drug increased comparable rates less when these occurred in the tandem DRL component than when they occurred in the tandem DRH components. Control rate is an important determinant of drug effects, independent DRH components. Control rate is an important determinant of drug effects, independent of reinforcement density maintaining rates, and independent of environmental influences, such as response-spacing requirements for food presentation, that can modify the extent of some drug-produced rate changes.     

Observational learning in monkeys

Observer monkeys were housed next to demonstrator monkeys that were conditioned to respond on a multiple reinforcement schedule whose components were fixed-ratio 32, variable-interval 3-min, and extinction 5-min followed by an additional 30 sec of extinction during which every response started a new 30-sec interval. After observational periods from 113 to 210 hr long, during which observers could not perform the response and were given no extrinsic reinforcers, their first-response latencies to fixed ratio and variable interval were as short as the demonstrators; and their rates of responding were well above pre-observational baseline levels. About 8 hr later, a temporal pattern of responding appropriate to the multiple schedule emerged, including non-emission of responses during extinction. Controls lacking the chance to observe did not develop typically patterned responding after 60 hr in one case and, in two other cases, after 80 hr during which, on two occasions, every one of 50 responses was reinforced. In a second experiment, the stimulus lights associated with fixed ratio and variable interval were presented simultaneously. Subjects chose one of the schedules by responding to one of the levers beneath the lights. All subjects initially chose fixed ratio. Seeing the demonstrators switch to variable interval, due to increases in the fixed-ratio requirement, had no effect upon observers, which continued to choose fixed ratio.     

Temporal properties of responding during stimuli that precede response-independent food

Pigeons' keypecks were reinforced with grain on the average of once per minute by schedules that maintained low response rates and by schedules that maintained high response rates. During these schedules, a fixed-duration conditioned stimulus (CS) ranging from 7.5 to 120 sec in duration across conditions terminated with response-independent food. Response rates during the CS were inversely related to CS duration. The rates and the temporal patterns of responding during the shortest CS were similar whether the ongoing schedule maintained high response rates or low response rates. As CS duration increased, the rate and pattern of responding during the CS converged on the rate and pattern of responding maintained by the baseline schedule. These data indicate that changes in responding during stimuli that signal response-independent reinforcement are not homogeneous throughout the CS; that response measures, such as “suppression ratios”, which presume homogeneity may mislead us; and that conditioned suppression and conditioned enhancement may be better talked about in terms of species-specific approach and avoidance than in terms of emotional states.     

SELF-MONITORING,REHEARSAL, AND DECEPTIVE COMMUNICATION

This study examined the effects of self-monitoring and rehearsal on the ability of observers to detect deception and on the behavioral correlates of deception. It was hypothesized that observers would be more accurate at detecting deception perpetrated by low self-monitors than by high self-monitors, with the difference particularly pronounced when messages were rehearsed. In addition, low self-monitors communicating spontaneously were expected to display greater rates of verbal and nonverbal responding than high self-monitors who planned their communications. Sixteen high and low self-monitors both lied and told the truth (either spontaneously or after 20-minute rehearsals) regarding their feelings while viewing slides of pleasant landscapes and of disfigured burn victims. Analysis of the responses of the 151 observers who made veracity judgments supported the hypothesis concerning accuracy of deception detection. Coding of 10 verbal and nonverbal behaviors revealed that unrehearsed low self-monitors displayed significantly greater pause and nonfluency rates than rehearsed high self-monitors. Additional findings are reported regarding the effects of self-monitoring, rehearsal, and truthful versus deceptive communication on the behavioral correlates of deception.     

A comparison of frequency, interval, and time-sampling methods of data collection

Data representing high, medium, and low response rates in constant and nonconstant patterns were generated by electromechanical equipment to determine whether the same data collected by time-sampling, interval recording, and frequency recording would be represented similarly by each method. Results indicated: (1) that time-sampling provided an extremely inaccurate estimate of responding, and (2) that interval recording accurately represented responding of low and medium rates, but grossly underestimated high-rate responding.     

Social influence in pigeons (Columba livia): the role of differential reinforcement

Socially-influenced learning was studied in observer pigeons that observed a demonstrator in an adjacent chamber performing a target response comprising standing on a box and pecking a key 10 times. In Experiment 1 there was no evidence for social learning in the absence of reinforcement of the observer's behavior. When the target response was already established in the observer's repertoire, but was not differentially reinforced in relation to the demonstrator's behavior, rates of extinction were not influenced by the demonstrator's behavior (Experiment 2). Reinforcement of the observer's target response in the presence of the modeled target response, and not in its absence, resulted in control of the observer's responding by the behavior of the demonstrator (Experiments 3 and 4). This control was extended in Experiment 5 to deferred responses that occurred following a delay since the demonstrator's target responses. The acquisition of social influence depended on differential reinforcement of the observer's target response, with the demonstrator's target behavior serving as the explicit discriminative stimulus.     

DIFFERENTIALLY REINFORCING LOW RATES OF MISBEHAVIOR WITH NORMAL ELEMENTARY SCHOOL CHILDREN1

Effective nonpunitive procedures for reducing counterproductive classroom behaviors are of potential benefit to both students and teachers. A recent strategy for dealing with this class of problem behaviors involves the reinforcement of acceptably low levels of such behavior. The laboratory version of this procedure, called differential reinforcement of low rates of responding (or DRL), provides for a reinforcer to be delivered contingent upon a response that is separated from the last preceding response by a minimum amount of time. To make this procedure more amenable to classroom use, the present authors have modified it so that a reinforcer is delivered if fewer than a specified number of responses occur within a preset time interval (Deitz and Repp, Journal of Applied Behavior Analysis, 1973, 6 , 457–463). Previous studies using this procedure have found it effective in reducing and maintaining low rates of targeted behaviors. However, these effects have been demonstrated with groups of subjects and/or individuals from dependent populations. The present study investigated use of this modified DRL procedure with individual students in normal elementary classrooms. In the first of three studies, “talk-outs” of an 11-yr-old fifth-grade male were reduced when nonexchangeable gold stars were made contingent on two or fewer responses per session. During baseline sessions, an average of 4.45 talkouts were observed per 45-min session. Average responding subsequently fell to 1.83 when the modified DRL contingency was applied, increased to 7.60 during a reversal phase, and dropped again to an average of 1.20 when the contingency was reapplied. In the second study, out-of-seat behavior of a 12-yr-old sixth-grade female was reduced when gold stars were made contingent on two or fewer responses per 45-min class period. Baseline responding averaged 6.10 responses per session. When the contingency was applied, average responding fell to 0.16. During the reversal period, responding increased to an average of 6.00 and fell again, after the contingency was re-introduced to an average of 0.40. In the third study, a reduction in both talking-out and out-of-seat behaviors of another 11-yr-old fifth-grade male was demonstrated with a multiple-baseline design. Using different lengths of baselines, gold stars were made contingent first on a low rate of out-of-seat behavior, and then on a low rate of talk-outs. Out-of-seat responding fell from a baseline average of 7.50 to a treatment average of 1.14. Talk-outs went from a baseline average of 4.66 to a treatment average of 1.14. In all three studies, the modified DRL procedure proved effective with the children and was manageable by the classroom teacher. For the students, nonexchangeable conditioned reinforcers (stars) were sufficient to maintain lowered rates of inappropriate behavior with the modified DRL schedule; there was no need for an elaborate token economy, a process that in many cases may be only a form of behavioral “overkill”. As in other studies investigating DRL schedules, students were not informed of their accumulation of responses; the differential effects of providing or withholding this feedback need to be investigated. Overall, these studies add single-subject replication with normal children to the literature on modified DRL procedures.     

Reactivity to home observation: a comparison of audio recorded behavior with observers present or absent

THE STUDY COMPARED THE BEHAVIOR OF FAMILY MEMBERS AT HOME AS RECORDED BY AUDIOTAPE RECORDINGS IN TWO CONDITIONS: with an observer present or absent. Behavioral differences were expected as a function of differential reactivity to these observational procedures, but none was found, and there was no evidence of adaptation effects in either condition. In general, significant positive correlations were obtained between the rates of recorded behavior in both situations. The implications of these findings for the development of nonreactive observation procedures were discussed.     

Effects of d-amphetamine on observing behavior in the squirrel monkey

Four squirrel monkeys were trained to press a lever, which produced stimuli indicating availability or non-availability of reinforcement for pushing a key. Food reinforcements were available for the key response at random intervals with an average rate of 1 per min. When food was available, a single lever response produced a red light behind the key. Reinforcement availabilities and red keylights remained until terminated by a reinforced key response. When reinforcement was not available, each lever response produced a 0.5-sec green light on the key. Except after lever responses, the key remained dark. Under this procedure, lever responses functioned as observing behavior in that they produced discriminative stimuli correlated with the availability or non-availability of reinforcement for key responses. The procedure generated a high rate of responding on the lever, short latencies of the key response after onset of red lights and few responses to the key in the absence of red lights. Intra-muscular d-amphetamine, in doses from 0.125 to 1.0 mg/kg, abolished both observing behavior and key responding for periods that increased as a function of dose. However, both observing and key rates were increased at the smallest dose in two subjects whose performances included responding to the key in the absence of red lights. Results are discussed in relation to previous findings regarding effects of amphetamines on operant behavior and on observing and monitoring performance.