Affect in the context of language: a bilingual experience

The purpose of this study was to determine whether language serves as a cue to affective states of bilingual individuals. Forty female coordinate Spanish-English bilinguals served as paid subjects. Velten's mood induction procedure (VMIP) was used both in its original form and in its Spanish translated form to induce mild, transitory mood states of depression and elation. The hypotheses of the study were: 1) significant differences in mood would be found between induced elation subjects and induced depression subjects, 2) the mood inductions in Spanish would have a greater effect than the mood inductions in English, and 3) greater effects for the mood induction would be found on a measure given in the same language as the mood induction than on a measure given in the other language. The independent variables were the mood of the induction (elation or depression), the language in which the mood induction took place (Spanish or English), and the language of the verbal dependent measure (Spanish or English). The mood of the induction and the language of the induction were between-subjects variables, whereas the language of the dependent measure was a within-subjects variable. The dependent variables were Spanish and English versions of the Multiple Affect Adjective Check List (MAACL) and a writing speed task. The MAACL is a verbal measure of mood, whereas the writing speed task is a behavioral measure of mood. Reading proficiency scores in each language were intended to serve as covariates; however, these scores failed to meet basic assumptions of covariance. A repeated measures analysis of variance was conducted for the MAACLs. An analysis of variance was conducted for the writing speed task. Results showed that the VMIP produced the expected changes in subjects' mood on the MAACLs but not on the writing speed task. However, neither the interaction between the language of induction and mood of induction nor the interaction between the language of induction, mood of induction, and language of MAACL were statistically significant. Therefore, only the first of the hypotheses was supported. Unexpected findings indicated that Spanish proficiency and the Spanish MAACL were positively correlated for the Spanish depression induction condition and that English proficiency and the English MAACL were negatively correlated for the English elation induction condition. These results are discussed. It is suggested that English may serve as a mediator to elated states and that Spanish may serve as a mediator to depressed states for subjects who are highly proficient in each of these languages. Spanish translation studies of the VMIP and the MAACL are presented in the appendices.