We are currently examining the effects of various psychoactive drugs on the social behaviors of individuals in a group of six stumptail macaques (Macacca arctoides) (Wilson, Bellarosa, & Bedford, 1977). Existing observational data systems were not adequate for our situation, primarily because they (1) did not permit the direct use of a system having high-speed file access, high-level languages, and sophisticated statistical packages (Fitzpatrick, 1977); (2) presented the observer with lengthy and arbitrary behavior encoding schemes (Sykes, 1977) that could not easily be learned due to the large number of behaviors we continuously sample; (3) did not permit the storage of enough data to recreate an entire observation session (Conger & McLeod, 1977). Our system must retrieve data pertaining to any environmental condition, any dyadic behavioral interactions (who interacts with whom), the duration of particular behavioral episodes, and the temporal relationship between any two behavioral episodes, and thus permit flexibility in the establishment of new data analysis routines. Users must be able to define new behaviors, delete old behaviors, and add observers. An encoding scheme using brief mnemonics was preferable since observers could more easily master the system, enter more behaviors per unit time, and minimize the chances of error. SOBEX Data Logging. To begin an observation session, the observer types in a sentence to the SOBEX monitor (described below) containing the key word, "CREATE" (e.g., "Create a data module"). The SOBEX monitor then runs the SOCIAL subprogram, which requests environmental data (observer name, drug, dose, numbers of drugged animals, room temperature) from the observer and outputs the information as the first five words of the data file. After an appropriate period designed to allow the onset of the drug effect, feeding order data is requested. The number of the animal taking each of 72 monkey chows is entered in the order in which the event occurs. After the last chow is removed, the observer types a "G" to start the session. When the SOCIAL subprogram "sees" a G command,
[1]
Lawrence J. Fitzpatrick.
BEHAVE—An automated data analysis system for observed events
,
1977
.
[2]
Marvin C. Wilson,et al.
Sociopharmacology of d-amphetamine in Macaca arctoides
,
1980,
Pharmacology Biochemistry and Behavior.
[3]
Richard E. Sykes,et al.
Techniques of data collection and reduction in systematic field observation
,
1977
.
[4]
Rand D. Conger,et al.
Describing behavior in small groups with the Datamyte event recorder
,
1977
.
[5]
Second Edition,et al.
Statistical Package for the Social Sciences
,
1970
.