Adult empathy and emotional availability are central to the healthy development of infants, yet there is a paucity of empirically based measures to assess these constructs. To address this need, a study was conducted to develop the Pictures of Adult-Infant Relations (PAIR) Test, and to determine its reliability and validity. The PAIR Test adapted Thematic Apperception Test methodology, using eight drawings of adults and infants. Empathy was defined as accurate perception of emotional state, and emotional availability as attentiveness and responsiveness to the infant.

Content validity was determined by a multidisciplinary panel of 31 expert judges who completed two questionnaires, evaluating the PAIR pictures' portrayal of emotional expressions and caregiver-infant interactions. Results indicated that the pictures portrayed both a range and balance of emotional states. Expert agreement also identified the degree of attentiveness and responsiveness shown by caregivers in each picture. These results provided scoring criteria for scales of Empathy and Emotional Availability.

Concurrent validity and reliability were tested with a sample of 25 mothers of young infants. The mothers ranged in age from 17 to 47; most were married, Caucasian, with some college education and adequate income. They completed the PAIR Test, as well as Davis' Interpersonal Reactivity Index, a standardized empathy questionnaire. Two observational measures relevant to emotional availability were also utilized: Ainsworth's First Quarter Scales of Maternal Care Behavior, and Caldwell and Bradley's Home Observation for Measurement of the Environment. Nine mothers completed the PAIR Test twice, with a two week interval. Results supported the conclusion that mothers' emotional availability as assessed through the PAIR Test was correlated with observational measures of their emotional availability to their own infants. However, the empathy scale of the PAIR Test showed no relation to the concurrent empathy measure. There were no significant differences between test and retest scores. Results pertaining to internal consistency within scale and subscale scores suggested that the PAIR Test may be measuring a single broad construct reflecting an individual's overall capacity for caregiving with infants.