AbstractNegotiations were conducted to investigate the effects on settlement‐points andon the attitudes and perceptions of participants of (i) group participation and (ii)belief in own group's point of view, in a 2 × 2 factorial design. Ninety‐six school children prepared cases in groups of four before representing their group's position against an individual of a similarly prepared opposed group. Group participation was manipulated by groups either participating in preparatory discussions or observing video films of another group's discussions. Belief was manipulated by systematically varying the composition of groups according to scores on a pre‐test of attitudes towards the raising of the school‐leaving age.In general the belief manipulation operated as expected, ‘believers’ exhibiting less variability, more tit‐for‐tat agreements and less opinion change than the ‘disbelievers’. Group participation did not influence the measures as predicted, and measures of interpersonal perception did not conform to the pattern of findings in recent experiments on intergroup discrimination. The results are discussed in terms of (i) their relevance to the issue of the appropriate relationship of the representative to his group in a negotiation and (ii) their implications for intergr