Meanwhile, in a very different arena, John B. Watson, the father of American behaviorism, blamed adult abnormalities on fears learned in childhood. Like Freud, Watson and his colleagues illustrated their theory with case histories of children's fears. In the case of Albert, for example, Watson and Rayner ( 1920) demonstrated the conditioning of fears to innocuous stimuli, whereas in the case of Peter, Jones ( 1924) demonstrated the deconditioning of fear.

Fears were also of considerable interest to students of normal child development, such as Jersild and Holmes, whose 1935 survey of children's fears became a classic of the child development literature. When efforts were made to combine psychoanalytic and learning theories in the 1940s and 1950s, the hypothesized childhood roots of neurotic anxiety retained a major role. Dollard and Miller ( 1950), for example, viewed anxiety as fear whose source is vague or obscured by repression. Such anxiety occurs when the victim lacks appropriate verbal labels, either because a fear began before language skills were advanced enough to provide labels or because repression separated verbal labels from the source of the fear.

With the rise of interest in personality traits, children's versions of adult anxiety tests were developed. The Children's Manifest Anxiety scale (CMAS; Castaneda, McCandless, & Palermo, 1956), for example, is a downward extension of the Taylor ( 1951) Manifest Anxiety Scale for adults. (A revision of the CMAS has since been published by Reynolds and Richmond, 1978, with the title "What I Think and Feel.") Dating from the same era as the original CMAS is the General Anxiety Scale for Children (GASC; Sarason, Davidson, Lighthall, Waite, & Ruebush, 1960). A specialized version of the GASC, the Test Anxiety Scale for Children (TASC), was designed to identify children whose anxiety about testing may interfere with their school performance ( Sarason et al., 1960). A more elaborate measure derived from adult tests is the State-Trait Anxiety Inventory for Children (STAIC; Spielberger, 1973), which has separate scales for the child's current anxiety state, as well as a more enduring anxiety trait. The Personality Inventory for Children (PIC; Wirt, Lachar, Klinedinst, & Seat, 1977), a multidimensional personality inventory completed by parents and modeled on the Minnesota Multiphasic Personality Inventory (MMPI), also contains an anxiety scale.

The revival of behavioral therapies in the 1960s rekindled interest in the direct observation and conditioning of fears. Behavioral treatments were illustrated in numerous case studies of adult fears, as well as in some child clinical cases and in experiments with children having subclinical fears. Rejecting "traditional" clinical assessment in favor of "behavioral" assessment, behavior modifiers devised their own procedures for documenting fearful behavior. These included behavioral avoidance tests in which the child is asked to perform a graded series of approach behaviors toward the phobic object (e.g., Bandura & Menlove, 1968; Kornhaber & Schroeder, 1975). Observational schedules have also been developed to assess signs of anxiety, such as crying, trembling, stuttering, and talking about fears (e.g., Glennon & Weisz, 1978; Melamed & Siegel, 1975). A behaviorally oriented self-report measure, the Fear Survey Schedule for Children ( Scherer & Nakamura, 1968), has been developed as a childhood version of Wolpe and Lang ( 1964) Fear Survey Schedule for adults. . .