A SYSTEMS APPROACH TO BEHAVIOR III: ORGANISMIC PACE AND COMPLEXITY IN TIME-SPACE FIELDS   ROGER D. RAY and JAMES D. UPSON Rollins College B. J. HENDERSON Florida Technological University

[We owe a great deal to colleagues who helped gather and analyze data upon which this paper is based. Special thanks are due Douglas A. Brown for work on the discriminative pacing experiment and to Merrill J. White, Jr., and Vicki Coleman for untiring contributions to the whale research. The micro-behavioral analysis project on the killer whale was supported financially in part by a grant from Hubbs-Sea World Research Institute; however, the material set forth herein does not necessarily reflect the views or conclusions of Sea World. We are most appreciative to Sea World, Inc., and especially to its Orlando (Fla.) staff for their cooperation in making these observations possible. Reprint requests may be sent to Roger D. Ray, Department of Behavioral Science, Rollins College, Winter Park, Florida 32789.]

 

This paper is the third in a series covering a systems approach to environmental-organismic interactions. Two experiments are presented which explore behavioral organization dynamics in temporally defined settings. The first study investigates relations between the change rate of imposed environmental settings and consequent changes in behavioral flow dynamics in laboratory rats. The second investigation focuses on biological rhythms in behavioral-respiratory dynamics of killer whales in oceanaria. By analyzing behavioral organization at both structural (micro) and functional (macro) levels, sequentially organized and recurrent behavior patterns were found to occur in temporally and spatially defined settings. Conclusions focus on the implications of reported data relative to: (a) requirements that adequate behavioral measurement include frequency, duration, and patterned integration; (b) systemics of psychological-physiological (organismic) integration; and (c) potential definitions of functional psychological time boundaries.

In 1975, Ray and Brown introduced a systems model for behavioral analysis which stressed that organisms and their immediate environments are constantly interacting with one another. These interactions take place within both environmentally and organismically bound setting conditions which often modify the dynamics of the interaction. A flow diagram (Ray & Brown, 1975) depicts these "interbehavioral" (Kantor, 1959) events as follows:

Illustrated behaviors (any Bn) are constantly interactive with specific environmental events (any SO, but are occurring with differing frequencies and durations.

Assessing the sequential organizational patterns and the rate of change in these events within blocks of time, Ray and Brown (1975) generated behavioral flow analyses (i.e., flow rate and sequential pattern variability) which demonstrated behavioral susceptibility to manipulations of the character of environmental (E s) and organismic (0 s) setting factors. Setting factors empirically probed by Ray and Brown included ambient chamber temperature, reinforcement deprivation operations, discriminative cue conditions, and pharmacological organismic states. But temporal and spatial organizations implied within and between general setting conditions were not explored in that paper.

The development of appropriate descriptive behavioral categories used to depict behavioral flow was mentioned by Ray and Brown (1975), but they did not point out that many levels of categorical analysis are possible. Thus, any given interbehavioral state (Sn-Bn-Sx) may be thought of as comprising multiple, and possibly recurrent, interbehavioral transactions which make up coherent micro structures within the more macro interbehavioral unit. A diagram depicting only the "nested" behavioral units (Ray, Upson & Henderson, in press) is as follows:

As illustrated in this flow diagram, behavioral categories may be chosen arbitrarily to depict either macro or micro organizational dynamics. Presumably, any coherent macro state could evolve empirically as assessed high- probability clusters of recurrent micro categories. Also, the most quantal or molecular level of analysis might well parallel the "perceptual moment," as envisioned by researchers interested in psychological time perception (cf. Cohen, 1967; Efron, 1972; Poppel, 1972).

As Schaltenbrand (1974) has suggested, time-space fields are fundamental elements of all natural systems. Thus, the temporal and spatial qualities of environmental setting conditions may be expected to have significant influence over interbehavioral parametrics. Unfortunately, research currently available on behavioral systems has failed to give these temporal and spatial elements due emphasis. The existence of such field forces in naturalistic settings should be amenable to clarification by continued application of the analog experimental approach, which stresses the continuity between laboratory and field investigations. The following two experiments were designed to elaborate further this design continuity.

EXPERIMENT 1: THE PACE OF CHANGE

IN ENVIRONMENTAL SETTINGS*

[*Research covered in this section was first reported at the 1974 annual meeting of the Pavlovian Society, Little Rock. The paper was coauthored and titled: Ray, R. D., Brown, D. A., & Greenspan, J. D. "Cardiovascular behavioral relationship changes in rapid paced environments."]

This experiment investigated the impact of changing durational parameters of environmental setting stimuli on behavioral organization and flow dynamics. In this analysis we have continued to apply a systems methodology in relatively standard animal research paradigms involving stimulus discrimination by bar-press and yoked-control groups of rats (Ray & Brown, 1975, 1976). Discrimination training establishes differentially important setting stimuli which may be temporally paced by sequential alternation.

Because yoked-control animals do not bar-press in positive discriminative stimulus settings, they represent different levels of effort expenditure than bar-pressing animals, and thus may serve as a control for potential differential effort and rate effects. Since interbehavioral flow dynamics are assessed in part by temporal measures, it is relevant that operant bar-press paradigms involve internal subject-paced interbehavioral transactions, while yoked- controls are essentially externally or other-paced. Interactions between environmental pacing variables and behavioral-organismic pacing variables thus are evaluated from stimulus discrimination pacing paradigms by including bar-press and yoked groups.

Four sets of eight 15-min. experimental sessions involving successive S+ and S- presentations were used. During the first and fourth set, durations were identical to prior discrimination training in which presentations of S + occurred 37 sec. apart (mean onset-to-onset) with a range of 15-85 sec. (S + or S- mean duration = 18.5 sec.; rapid pace). During the second set of conditions (medium pace), S + and S- durations were twice as long as those of rapid pace (mean= 37 sec.). A third condition (slow pace) included stimulus durations twice those of medium pace, or four times longer than those of rapid pace (mean= 74 sec.). Multiple behavior category measures are defined elsewhere, along with a detailed rationale for their application (Ray & Brown, 1975).

Figure 2. (A) Mean frequency of behavior-to-behavior shifts/min. (behavioral flow rate) occurring within each stimulus condition across all successive sessions and schedule pace conditions for response-contingent (RC) and free reinforcement (RF) groups. (B) Mean number of different types of behavior shifts from one category to another (sequential pattern variations) occurring within each stimulus condition across all successive sessions and schedule pace conditions for response-contingent (RC) and free reinforcement (RF) groups.

A sequential analysis of behavior category changes allows variations in counts of (a) total numbers of behavioral changes within a specified setting and time period (behavioral flow rate) and (b) the different kinds of behavioral shifts, or sequential patterns, within a specified setting and time period (sequential pattern variations). Systematic differences between experimental groups, between S + and S- situations, and between stimulus durations, or pacing manipulations, were found in both of these measures (see Figure 2).

Likewise, as stimulus setting schedules were lengthened, the number of variations in types of behavioral sequencing patterns during S- (Figure 2b) progressively was reduced from initial rises for both experimental groups. Reinstatement of the rapid pace condition returned behavioral sequencing rates and pattern variations to their original higher levels for S- periods in both groups. The relatively small degree of S+ pattern variation change and the lack of S + flow rate change apparently relates to the constancy of reinforcement rate in both groups; bar-press rates, and thus reinforcement rates for all animals, remained unaffected by stimulus duration changes. Task requirements, in the form of bar-presses, are not critical to setting schedule effects, since the effects were observed in both response-contingent (i.e., task) and free reinforcement (i.e., non-task) groups.

The oscillatory character of S- declines in behavioral flow rates illustrated in Figure 3 is highly reminiscent of Lat's (1966; Lat & Gollova-Hemon, 1969) ingenious work on the damped oscillations inherent in habituation/adaptation/extinction phenomena. It is noteworthy that Lat's measures of nonspecific excitability level (NEL) and lability, used to assess adaptational oscillations, were derived from techniques quite compatible with those used in the present study. Our results suggest that shifts from one environmental setting condition to another (i.e., from S + to S-) induce adaptational requirements in behavioral flow rate throughout each and every exposure to that setting shift. Furthermore, the temporal-durational parameters attending the preceding S + setting condition seem to affect both the initial rate levels from which behavioral flow rate adaptations in S- begin and also the damping constants involved in the oscillatory decay function (e.g., Graupe, 1972).

Figure 3. Mean changes in behavioral flow rate during each successive 5-sec. block following Sonset and continuing until the mean S+ onset. Data are from the response-contingent (RC) group only, and depict changes during the combined final two sessions of rapid S + /S- pace and slow S + /S- pace, respectively.

Finally, oscillatory behavioral flow dynamics also seem to exist in response to shifts in setting pace. When such a shift first occurs, as from medium pace to slow pace, behavioral flow dynamics initially are elevated in both rate and variability; however, across sessions they gradually settle to levels appropriate to specific setting durations. Further research is needed to determine if these across-sessions rate and variability declines are truly oscillatory in nature, as the S- data in Figures 2a and 2b suggest strongly. Thus, in each pacing condition, the fifth, sixth, or seventh session always represented a phasic "second peak" in measurement levels. These multiple declines in variability and behavioral flow rate, both within each individual S- period and within each general pacing condition, strongly parallel Lat's (1966; LAt & Gollova-Hemon, 1969) observations of within-session and across-sessions oscillation families. Systematic adaptational/habituational dynamics relating to general interbehavioral organization are indicated in the present study.

Conclusion

It is relatively well-established that temporally paced punctate stimuli (e.g., reinforcers) serve to organize behavior across time (Killeen, 1975; Schoenfeld & Cole, 1972). Lat's work (1966; Lat & Gollova-Hemon, 1969) on habituation relied on changing the animal's location from one setting to another, and thus was among the first demonstrations that temporally ordered behavioral events stem from more general contextual setting changes. The present study goes further to demonstrate that not only changes in spatial settings but also the pace of meaningful stimulus fluctuation has an impact on the temporal structure of behavioral flow.

It is important to note that the time frames used in these various investigations of temporal-environmental setting events are very short compared to those time periods normally investigated relative to temporal-organismic setting factors, or biological rhythms. Under natural conditions, the earth's rotation marks a 24-hr. cycle which is partitioned into alternating periods of light and darkness. Most animals, if not all, begin and/or cease activity according to such demarcations as twilight (known as zeitgebers) and thus are organized around a circadian (i.e., approximately 24-hr.) cycle of activity fluctuation (Kavanau & Peters, 1976; Luce, 1971; Scheving, Halberg & Pauly, 1974). Other astronomical time scales, including lunar and annual rotations, are related to biological functioning as well (Richter, 1974). Even without specific exogenous stimuli such as light and dark, animals in artificially constant environments continue to "free run" in endogenously determined cycles of only slightly different periods, Unfortunately for our present interests, researchers have yet to focus on specific behavioral details relative to such biological rhythms.

Our present work has led us to ponder these broader temporal boundaries of behavioral oscillations. If behavioral organization reflects oscillatory reactions to short time-framed stimulus perturbations, would other forms of "background" behavioral oscillations appear in broader time-framed perspectives? Noting that certain autonomic functions (e.g., the heart beat) demonstrate both short-term oscillations (e.g., sinus arrhythmia) and circadian oscillations (Luce, 1971), and also noting Folk's (1974) suggestion of coupling between somatic and autonomic activities on a circadian scale, we sought to discover whether specific behavioral oscillations could be found for circadian periods as well as our short-lived periods. Further, we explored potential relations between circadian fluctuations in behavioral states and concomitant physiological functions.

Next section..... EXPERIMENT 2: THE CIRCADIAN STRUCTURE OF BEHAVIOR AND RESPIRATION