”The care with online privacy is a matter of social responsibility.” (Novaes Neto, 2011)
Neto, N. N. (2011). The Internet as an experimental laboratory for behavior analysis. Master’s Thesis Postgraduate Program of Studies in Experimental Psychology: Behavior Analysis. 88 p. PUC-SP.
Thesis Advisor: Sergio Vasconcelos de Luna
Line of Research: Basic Processes in Behavior Analysis
ABSTRACT
The present research had as primary goals: 1) To present a proposal for experimental procedure involving the use of the Internet as a laboratory for analysis of human behavior. 2) To use an online chat environment in order to assess whether users’ behavior of switching between chat rooms in a chat service can be influenced by the introduction or removal of a stimulus (CAPTCHA). A total of 700 users participated in an experimental design (three phases) to verify whether the manipulation of the CAPTCHA stimulus in stages 2 and 3 would affect users’ behavior as they enter the chat rooms. Two measures were employed to measure behavioral changes: a) number of times each user entered a chat room; and b) amount of time daily spent in the chat rooms for each user. It was hypothesised that for a considerable number of users, the removal of the CAPTCHA would increase the chances of switching to another chat room because of the elimination of the nuisance of answering the test. The results indicated that the pattern of behavior of users who entered the chat rooms was influenced by the effect of the manipulation of the CAPTCHA variable. However, the results were not significant enough to confirm the hypothesis. The measures employed to assess behavior changes were shown to be statistically significant when the CAPTCHA variable was reintroduced (Phase 3) to the participants after a 15-day period without the presence of the independent variable (Phase 2). This effect has shown that the reintroduction of CAPTCHA to the chain of gaining access to the room has significantly affected the behavioral pattern of the users group, as compared to the previous stages, indicating that the process of reintroducing VI had, in part, aversive properties. All the same, it is important to consider that it has not been possible to determine whether this effect was caused by the influence of the CAPTCHA stimulus or by the simple change of environment with the inclusion of another component and condition in the chain of gaining access to the room. Since more and more people are increasingly using the Internet, the study of this environment can make a strong contribution to the science of behavior analysis.
Key words: CAPTCHA, social network, aversive stimulus, online privacy, chat.
Presentation (portuguese)
