“…The present study also emphasizes the importance of sender effects This IS compatible with otber applications of the social relations model to the analysis of dyads For example, m a reanalysis ofthe Lanzetta and Kleck (1970) study of nonverbal communication accuracy, Kenny and La Voie (1984) found 49% ofthe variance to be attributable to sender effects and 51% to tbe unique relationsbip and/or error No variance was attnbutable to receiver effects Similarly, m a reanalysis of a study employing the encoding and decodmg of posed expressions (Goldenthal, 1981), Kenny and La Voie (1984) found 44% of the variance attributable to sender effects, 2% to receiver effects and 54% to tbe unique relationship and/or error Kenny and La Voie (1984) point out that, although the social relations model suggests that sender effects are critically important, there has been relatively little research on what makes people good senders The available evidence suggests that sending accuracy is based upon a combination of temperamental factors and social learning (Buck, 1979) Also, there is evidence that individual differences m sending accuracy are related to psychophysiological responding m a manner descnbed by the externalizmg-mtemalizing distinction 0ones, 1950) expressive persons tend to have smaller skm conductance and heart rate responses to emotionally loaded slides and shocks than do nonexpressive persons (Buck, 1979, Buck et al , 1974, Lanzetta and Kleck, 1970 This has important lmphcations for behavioral medicine, for there is increasing evidence that an inability to express emotion (termed alexithymia, or "no words for mood") is associated with the etiologv of a vanety of stress-related diseases (cf Sifheos, 1983, Anderson, 1981…”