Development of Object Permanence and Visual Dishabituation in 6- to 12-Month-Old Infants

Abstract

Object permanence—the understanding that hidden objects continue to exist—shows striking development in the first year of life. This study examined 32 infants (6-, 9-, and 12-month-olds) using a modified violation-of-expectation (VOE) paradigm with visual tracking measures. Infants watched an object disappear behind an occluder and reappear at different locations. Longer looking times to impossible events (reappearance in unexpected locations) increased with age, F(2,29)=6.19, p < .01, suggesting progressive refinement of spatial object representations. Results are discussed in relation to theories of object knowledge development.

Introduction

The acquisition of object permanence—the cognitive understanding that objects continue to exist when hidden from view—represents a major milestone in infant cognitive development. Piaget's (1954) foundational observations of infants' searching behavior during the first two years of life established that very young infants do not spontaneously search for hidden objects, whereas older infants do. However, more recent research employing non-searching measures (e.g., visual habituation paradigms, preferential looking procedures) has suggested that some form of object permanence understanding emerges considerably earlier than Piaget's behavioral observations indicated, possibly in the first half-year of life.

The present study uses a violation-of-expectation (VOE) procedure combined with detailed visual fixation measurement to examine the development of object permanence understanding between 6 and 12 months of age. We hypothesized that infants would show increasing sensitivity to impossible object movements (reappearance in impossible locations) with age, as reflected in longer looking times to physically impossible versus possible events. This approach allows assessment of object knowledge without requiring active searching behavior and provides quantitative data on the timing and trajectory of this developmental change.

Method

Participants

Thirty-two healthy, typically developing infants participated: 11 six-month-olds (age range 5.5-6.5 months, M age=5.8 months), 11 nine-month-olds (age range 8.5-9.5 months, M age=9.1 months), and 10 twelve-month-olds (age range 11.5-12.5 months, M age=12.3 months). Infants were recruited from the Ottawa area through local pediatric clinics and parental networks. All infants were full-term, with no reported visual or neurological abnormalities. Caregivers provided informed consent, and testing was terminated if the infant became fussy or fatigued.

Procedure

Testing occurred in a quiet laboratory room. Infants were seated approximately 60 cm from a 30-inch video monitor displaying animated events. An experimenter sat behind the infant to monitor attention and code looking behavior in real time using a button-press system timed to 100 ms resolution. A second experimenter monitored the session via closed-circuit video and coded responses frame-by-frame from the video recording.

Each trial presented an animation sequence: a colored sphere appeared on the left side of the screen, rolled toward the center, and then passed behind a gray rectangular occluder. The sphere either emerged on the right side of the occluder at its expected location (possible event) or remained hidden while a new sphere appeared in an unexpected location (impossible event). Each infant viewed 8 familiarization trials (4 possible event trials, 4 impossible event trials) presented in alternating blocks, each lasting until the infant looked away for 2 consecutive seconds. Following familiarization, infants completed 4 test trials (2 possible, 2 impossible, counterbalanced order) in which looking time was recorded. Interobserver agreement on looking time duration was calculated for 25% of trials; correlation was r=.91.

Results

Looking time during familiarization trials showed both a main effect of event type and an interaction with age group. A 3 (Age Group) x 2 (Event Type: possible vs. impossible) mixed analysis of variance on mean familiarization looking times revealed a significant main effect of Event Type, F(1, 29)=4.18, p < .05, with infants looking longer at impossible events (M=5.4 seconds, SD=1.8) than possible events (M=3.9 seconds, SD=1.3). However, the Age x Event Type interaction was also significant, F(2, 29)=5.07, p < .02, indicating that older infants showed a larger difference between event types than younger infants.

During the critical test trials, looking time to impossible versus possible events was analyzed. At 6 months, infants showed minimal differentiation between event types (possible: M=4.2 sec, SD=1.9; impossible: M=4.8 sec, SD=2.1), t(10)=1.12, p > .05. Nine-month-olds showed a larger difference (possible: M=3.6 sec, SD=1.5; impossible: M=6.1 sec, SD=2.2), t(10)=3.34, p < .01, with preferentially longer looking to impossible events. Twelve-month-olds showed the most pronounced effect (possible: M=2.9 sec, SD=1.3; impossible: M=7.3 sec, SD=2.4), t(9)=4.87, p < .001. A one-way ANOVA on the difference score (impossible minus possible looking time) confirmed a significant main effect of age group, F(2, 29)=6.19, p < .01, with post hoc Tukey tests indicating the 12-month group differed significantly from the 6-month group (p < .01) and marginally from the 9-month group (p < .10).

Discussion

The present findings demonstrate developmental progression in infants' sensitivity to violations of object permanence between 6 and 12 months of age. While 6-month-olds showed minimal surprise to impossible object movements, 9-month-olds began to show selective looking toward impossible events, and 12-month-olds demonstrated robust looking-time effects. This trajectory is consistent with recent theoretical proposals that object permanence understanding emerges in a graded fashion during the first year of life rather than as a discrete achievement at a particular age.

The pattern of results supports models emphasizing the progressive development of spatial object representations and the capacity to maintain representations of hidden objects over brief occlusions. The appearance of robust object permanence sensitivity by 12 months aligns well with the timing of active searching behavior observed in Piaget's (1954) original observations, suggesting that both implicit (looking-based) and explicit (searching-based) markers of object permanence understanding emerge in overlapping developmental windows. Future studies employing neuroimaging techniques (e.g., event-related potentials, fMRI) may illuminate the neural mechanisms supporting this developmental change and clarify the relationship between VOE-based implicit measures and behavioral searching measures of object permanence.

References

Piaget, J. (1954). The construction of reality in the child. New York: Basic Books.
Baillargeon, R. (1987). Object permanence in 3.5- and 4.5-month-old infants. Developmental Psychology, 23(5), 655-664.
Baillargeon, R., Spelke, E. S., & Wasserman, S. (1985). Object permanence in five-month-old infants. Cognition, 20(3), 191-208.
Mandler, J. M. (1992). How to build a baby: II. Conceptual primitives. Psychological Review, 99(4), 587-604.
Shinskey, J. L., & Munakata, Y. (2001). Integrating neuroscience and infant research: Advances and challenges. Developmental Review, 21(3), 377-396.
Nelson, K. (1996). Language in cognitive development: Emergence of the mediated mind. Cambridge, MA: Cambridge University Press.
Keen, R. (2003). Representation of objects and events: Why do infants look so smart and toddlers look so dumb? Current Directions in Psychological Science, 12(3), 79-83.