An emerging body of research in group decision support systems (GDSS) provides evidence that computer technology can and does impact the quality of decision making in groups. Most GDSS research is oriented toward examining the effects of a computer system on group outcomes, typically decision quality or group consensus, with the process itself often treated as a "black box." The research reported in this article addresses the need for a closer, micro-level examination of group process. An important group variable, namely influence behavior, was isolated and examined at various levels and by multiple methods. A model of specific GDSS effects on influence behavior was developed, based on an information exchange view of decision making and on the impact of a GDSS as a communication channel. Based on the research questions of interest in the study, several propositions and hypotheses were advanced and empirically tested on a specific implementation of a GDSS. Results were analyzed both quantitatively and qualitatively. The major empirical findings of the study showed no significant difference between the overall amount of influence behavior attempted in computer-supported versus unsupported groups, although significant differences were found in the pattern of influence behaviors, i.e., the different types of behaviors used. In addition, the distribution of influence behavior was more even in GDSS groups than in unsupported groups on one of two measures used. Empirical findings partially supported the research model, with indications that decision-making groups need more active guidance in understanding how to adapt computer support technology to their view of decision-making processes.