Anthony James (Tony) DeCasper, Professor Emeritus of Developmental Psychology at the University of North Carolina, Greensboro.
Tony DeCasper passed away from complications of glioma at the age of 75 on July 4, 2016. He received his Bachelor’s degree from the University of Akron, served in the US Air Force from 1963-1967 (honorably discharged at the rank of Captain), began his graduate work in developmental psychology at the University of Iowa, and received his PhD from Emory University in 1974. He directed the Neonatal Perception Laboratory at the University of Kansas Medical Center from 1975-1976 before joining the faculty of UNCG and retiring in 2012. Recently, he made his home in Paris and Trouville.
Tony’s scientific career is marked by his ground-breaking research demonstrating how prenatal and neonatal auditory perceptual experiences contribute to the development of human cognitive abilities. He devised a research strategy that was able to reveal the prenatal mechanisms of learning and memory that establish the neonatal abilities essential for the continued development of the infant. Tony was one of only a few scientists studying human prenatal and newborn behavior in the 1980s.The hypotheses tested with the differential reinforcement sucking procedure that he designed, revealed the developing cognitive abilities of the perinatal human. During his initial work with newborns, he observed similarities in infant sucking and avian pecking behavior and developed a non-nutritive sucking procedure based on the burst-pause pattern typical of newborn sucking. In this procedure, infants heard an auditory stimulus contingent upon their sucking. However, the procedure afforded infants the opportunity to “choose” to hear one of two available stimuli by modifying the temporal properties of their sucking patterns. Tony observed that sucking burst durations are correlated with the preceding pause or inter-burst-interval (IBI) so that by presenting different stimuli contingent on production of shorter and longer pauses (IBIs), he could examine the relative reinforcing value of the two stimuli. Although this differential reinforcement or “choice” procedure was grounded in operant conditioning, Tony used this procedure to explore infants’ responsiveness to complex, naturalistic, auditory events, and these studies yielded new evidence for memory, selective attention, and the impact of prenatal experience on early perception, behavior, and cognitive development.
Tony’s methods provided the first direct experimental evidence that human newborns recognize the maternal voice (DeCasper & Fifer 1980), implying that the fetal brain encodes and retains some properties of the sound stimuli experienced in utero. However, definitive evidence for the effect of prenatal rather than postnatal experience on infants’ preference for the maternal voice could only be provided by manipulating in utero exposure to specific stimuli. Therefore, Tony completed a series of studies which supported the prenatal auditory learning hypothesis. This work demonstrated that newborns choose to hear or prefer: the mother’s voice compared to a strange female voice (DeCasper & Fifer 1980), a particular speech passage (DeCasper & Spence 1986) or melody (Panneton & DeCasper 1986) that the mother frequently recited or sang only while pregnant when compared to similar but novel stimuli, low pass filtered voices that simulate vocal properties transduced into the uterine environment versus the airborne postnatal voice (Spence & DeCasper 1987), and heartbeat sounds (Panneton & DeCasper 1984; Davis & DeCasper 1989). Significantly, these studies also revealed that newborns remember the maternal voice independent of the specific speech produced, and they recognize the prenatally experienced speech passage or melody even if it is produced by an unfamiliar speaker in the postnatal test. Consequently, both voice-specific and language-specific properties are encoded during the prenatal period and newborns are capable of retrieving these independently.
Other evidence for cognitive processing was provided by studies showing that newborns were sensitive to, and retained, very specific response-consequence contingencies or “rules” for accessing auditory stimuli over intervals as lengthy as 20 hours. For example, when newborns heard singing as a consequence of producing specific IBI patterns and were then retested 20 hours later, they retained and produced the sucking pattern learned previously, showing clear evidence for a kind of procedural memory. Additionally, some infants first experienced a session in which the auditory stimulus was presented independently of the infant’s sucking behavior, and then the session 20 hours later presented the same auditory stimulus contingent on production of the same temporal pattern that the infant produced in the first session. In this situation, infants failed to learn the response pattern necessary to hear the stimulus. That is, infants first learned that the auditory stimulus was independent of their behavior and they did not attend to or detect the contingent relation that existed for the second session (DeCasper & Carstens 1981). These studies demonstrate that newborn infants constructed expectations about their own behaviors, the auditory stimuli that are available to them, and the relationships between their behavior and the auditory consequences of their behavior, and that this “expectation” affected their attention to their environment. Thus, infants are learning that their behavior produces interesting sounds in their social environment and that they can control these sounds via their behavior.
Indeed, Tony’s procedure also revealed that newborns are sensitive to subtle temporal dimensions within the task; Tony and his students found that newborns’ “estimation” of IBIs reflects a Stevens-like exponential psychophysical function, while their discrimination between IBIs adheres to a Weber-like function (DeCasper 1980; DeCasper & Spence 1991) – evidence that directly connects this work with information processing programs designed for decision-making algorithms and robotics. Tony’s procedure afforded insight into the temporal discrimination abilities of newborns, the assessment of infants’ detection of different types of response-reward relationships, the evidence that newborns develop expectations about how their behavior affects the environment, and the long-term memory for these experiences.
Tony and his students also found that 2-3-day-old infants show an early ear asymmetry in the processing of sound. The characteristics of the sound to be learned interacts with the ear into which the infant chooses to direct it (i.e., when heard monaurally). Thus, rapid temporal variations (“clicks”) are learned when the left ear is stimulated, while longer-lasting spectral changes (prosody) are mainly learned when the right ear is stimulated (DeCasper & Prescott 2009). This discovery has important implications for the as yet unappreciated consequences of cochlear implant interventions and for understanding the development of hemispheric specialization of the brain for cognitive, language, and emotional function.
Importantly, Tony not only reported evidence for these capacities of human infants within the first two to three postnatal days, he extended these to the study of similar abilities in the late-gestation fetus. With a French CNRS group in Paris, he used fetal cardiac changes as evidence of fetal discrimination of different elementary sound categories, such as musical tones and syllables, and sound streams, such as short melodies and speech sequences in different languages (Lecanuet, Busnel, DeCasper et al. 1986; Lecanuet, Granier-Deferre, DeCasper et al. 1987; Lecanuet, Granier-Deferre, Jacquet & DeCasper 2000;Ribeiro, Granier-Deferre, Jacquet & DeCasper 2004). The fetuses displayed a cardiac orienting response only to the “familiar” speech sequence and this result was independent of the maternal voice that spoke it aloud during the exposition phase (DeCasper, Maugeais, Lecanuet, Granier-Deferre & Busnel 1986; DeCasper, Lecanuet, Maugeais, Granier-Deferre & Busnel 1994). Also, the cardiac orienting response was recorded at an unusually low intensity (SPL), indicating that it was below the autonomic response threshold, so the changes in heart beat is not simply a consequence of the stimuli activating the autonomic nervous system.
Later, Tony also engaged in research to determine how early in development such learning could be evidenced (Krueger, Holditch-Davis, Quint & DeCasper 2004) and how long a prenatal memory trace could last after birth (Granier-Deferre, Bassereau, Ribeiro, Jacquet & DeCasper 2011). The latter study showed that a melodic contour repeatedly experienced by human near-term fetuses elicits a strong cardiac orienting response one month after birth in sleeping infants. Both melodies used in the study only differed by their ascending vs. descending melodic contour. The control near-term fetuses and the control infants had similar cardiac response to the two melodies, indicating that they were not distinguished by the infants. The only group that showed a differential response to the melodies was the one previously exposed in utero. These results extended the retention interval over which a prenatally acquired memory of a specific sound stream can be observed from 3-4 postnatal days to up to six weeks. There is no comparable study on the duration of an auditory memory formed prenatally.
Obviously, these findings had an important impact on developmental science and pediatric practice. Tony provided the first empirical evidence that the capacity for learning and memory in newborns could be harnessed to reveal important perceptual proclivities that shape and guide early psychological abilities and their development. His findings contrasted with the widely-held view that perceptual abilities at birth constituted an “initial state” determined by biological factors and for which experience had little if any impact (DeCasper & Spence 1991). Instead, his work showed that experience begins to shape perceptual and cognitive functioning before birth and his experimental studies revised developmental scientists’ view of perinatal development from one of discontinuity between pre- and postnatal periods to a more accurate view of continuity during this period.
The findings that prenatal experience biases infants’ postnatal attention to previously heard speech and its specific properties revolutionized the research and theoretical perspectives in the field of language acquisition. The explosive growth of empirical studies into infants’ early perception of speech and its role in guiding language acquisition is testimony to how Tony’s ground-breaking experiments opened the door to a “revolution” in developmental science. Indeed, the impact on the literature of Tony’s 1986 paper demonstrating that newborns recognized the mother’s prenatally-produced speech has actually peaked since 2009, with nearly 30% of all citations of the 1986 paper occurring between 2009 and 2015. The 1980 DeCasper & Fifer paper has also continually influenced thinking in psychology, with over 2000 citations. Indeed, in a recent poll of the members of the Society for Research in Child Development, the International Society on Infant Studies, and the Cognitive Development Society, that 1980 paper was listed as the fourth most fascinating paper published in the period from 1960 to 2012. More importantly, Tony’s work has continued to affect a variety of research areas, including studies of attention, learning and memory; speech perception and language, ethology and psychobiology of development, neuroscience of brain plasticity, as well as philosophy of mind and anthropological theory on the earliest steps of enculturation.
The different strands of Tony’s research achievements have substantial implications for several key aspects of the earliest manifestations of human cognitive development. Research conducted with an alternate version of the operant choice procedure has shown that neonates can adjust or synchronize their oral motility to a sound stimulus to trigger a preferred auditory experience. Specifically, the learning sequence underlying the operant procedure established by Tony involves: 1) hearing and discriminating two tones; 2) associating each tone with a specific auditory stream; 3) retrieving in working memory the tone that will trigger a particular stream of speech sequence (involving recognition memory);4) selecting the best motor response among different movement possibilities (procedural memory); and finally 5) adjusting the motor response to a certain time interval. This indicates that human newborns can rapidly understand cause-effect relationships – or rules -, an ability that has been attributed only to older infants. Tony’s results also show that fetal/neonatal infants can rapidly create expectations about the environment which can be retained for a memory span of several weeks. Further, these results tap into processes related to proprioception and its multisensory linkages to time perception and social cognition, to the self-regulation of behavior as a function of a “goal” (hearing preferred sounds), to the earliest induction of rules and their translation into effective action, to the development of awareness of learned contingencies.
Tony’s approach also reveals the communicative behavior of infants in their everyday social niche. Early communication with caregivers is based on rules governing the associations between production and reception of sounds. Development entails increasing selectivity and synchronization between the performer and receiver, and a growing appreciation of which action to display in order to elicit the desired response from the receiver. This work has focused scientific attention on the altered pathways to normal communication and cognition associated with the externalized fetus, i.e., the premature infant. The revelation provided by Tony and his colleagues that the complex phenotype we refer to as cognition does in fact have prenatal roots, provided the backbone for the rapidly evolving field of developmental care for premature infants. Moreover, the recognition that the prenatal environment can profoundly affect normal development of multiple phenotypes, paired with the recent findings detailing fetal origins for vulnerability for childhood and adult disease, has been instrumental in focusing attention on this important period of development.
Finally, Tony’s work is highly relevant for the developmental neurocognitive understanding of speech perception. Human prenatal memory studies are interpretable in terms of enduring neural tuning processes, in line with data from animal (avian to mammal) models showing that early auditory experience modifies behavior as well as the anatomical-functional organization of the auditory nervous system, and coding properties at all levels of the auditory network. Early auditory experience induces cellular and synaptic plasticity with enhanced receptive field selectivity and attuned representations in the brainstem and primary auditory cortex for specific features of a recurring stimulus. The new generation of studies using fetal fMRI and MEG are now showing cortical neuronal activation in human fetuses and newborns (e.g., Jardri et al. 2012; Partanen et al. 2013), reflecting the very early perception-discrimination of voices and speech units. However, the behavioral demonstration, adaptive meaning, and cognitive consequence of these early neurobehavioral capacities were already established 30 years ago with the work of Tony DeCasper and his students. They established that the elementary mechanisms underlying human cognition develop progressively before birth, and significantly affect postnatal behaviors, in much the same way as in nonhuman species.
Tony is survived by his immediate family: daughters Anna (John) Lair and Hester (Brian Smith) DeCasper; treasured granddaughters Marie and Dorothea, and Helen Killacky; and his sisters Madeline (John) Vincent, Linda (Harry) Skeen and Mary Beth (Greg) Frohnapfel.