\ selected cro-'-scction of the working population (N = 642) wa^ interviewed with respect to their job motivations The extent to which extrinsic or intrinsic job components were valued wa;-found to be related to occupational level At higher occupational levels, intrinsic job components lopporiumt> for selfexpression, interest-value of work, etc) weie more \alued At lower occupational levels, extrinsic job components (pa\, securin, etc) were more valued No sex differences were found in the value placed on intrinsic or extrinsic factors in general Howe\ er, women placed a higher value on "good co-workers'' than did men, while men placed a relatively higher value on the opportunity to use their talent or skill.: : c

Acted, videotaped messages containing conflicting inputs (friendly or unfriendly) in verbal (content), vocal (tone), and visual (facial expression, etc.) channels were shown to children (N = 80) and their parents (N = 80). "Joking" messages (criticisms said with a smile) were interpreted more negatively by children than adults; this age difference was most pronounced for female joking. Additionally, there was a general difference in the ratings of conflicting messages from men and women: women were rated as more negative (unfriendly). A linear model was found to be inadequate in accounting for the integration of conflciting messages; a strong interaction was found between verbal and vocal channels (a positive input in one channel was discounted if the other channel was negative).This study had two central goals: (a) to compare the interpretation of conflicting communication by adults and children and (b) to compare the adequacy of a linear versus an interactive model in accounting for the interpretation of conflicting communication. Using brief videotaped messages as stimuli, the authors examined how contradictory information in verbal and nonverbal channels (e.g., a criticism accompanied by a smile) is differentially combined and interpreted by children and adults.Situations which simultaneously contain pleasant and unpleasant elements are inherently tension-inducing. If the conflict can be resolved in the direction of "decreased incongruity" and denial of the negative element, the ultimate consequence is likely to be tension reduction or relief. Humor is often characterized on this basis (Berlyne, 1969). Sarcasm and kidding are common examples of communications which give simultaneous approval and disapproval. But conflicting messages do not necessarily result in a positive 1 The authors wish to express their appreciation to Norman H. Anderson for his thoughtful and cogent suggestions with respect to interpretation and presentation of results.

An analysis was made of videotaped parent-child communication within 20 families containing a "disturbed" child (referred by schools for chronic behavior or emotional problems) and 10 "normal" control families. Parental messages were judged for evaluative content in verbal (typescript of message), vocal (tone of voice), and visual (facial expressions, gestures, etc.) channels. Significant!;' more (p = .05) disturbed mothers (59%) produced messages containing evaluative conflict between channels than did normal mothers (10%); no difference was found between normal and disturbed fathers, The sons of mothers producing conflicting messages were found to be higher on school aggressiveness than the sons of "noncon dieting" mothers (p = .05).

In this study the authors explored developmental trends in the perception of acted, video-taped messages which were systematically varied in the channel used to convey evaluative content. Messages were either positive (friendly), neutral, or negative (unfriendly) in the verbal channel (content), the vocal channel (voice tone), and/or the visual channel (facial expression, etc.). Middle-class children, aged 5-18 (N = 120), and their parents (N = 80) were compared for their ratings of these messages. An age trend was found only for the visual channel (which had a reduced impact on young children). This trend was strongest and most reliable (as tested in a replication) for the perception of women's smiles-seen as neutral or only weakly positive by young children (5-8). Additionally, information was obtained regarding integration of multichannel messages and differential perception of male and female encoders.

Ratings were made of videotaped verbal and nonverbal (smiling) behavior of parents in interaction with their children. The sample included 20 families containing a disturbed child and 20 normal control families. An interaction was predicted and confirmed (p = .05) between parent sex and facial expression; that is, when a father smiled, he was making a friendlier or more approving statement than when he was not smiling; for mothers, there was no difference in the evaluative content of verbal messages when she was smiling versus when she was not smiling. This pattern was found to be unrelated to child disturbance. The findings were, however, limited to middle-class families because the majority of lower-class mothers in this sample did not smile at all.