Fusion of visual cues is not mandatory in children

Abstract

Human adults can go beyond the limits of individual sensory systems' resolutions by integrating multiple estimates (e.g., vision and touch) to reduce uncertainty. Little is known about how this ability develops. Although some multisensory abilities are present from early infancy, it is not until age ≥8 y that children use multiple modalities to reduce sensory uncertainty. Here we show that uncertainty reduction by sensory integration does not emerge until 12 y even within the single modality of vision, in judgments of surface slant based on stereoscopic and texture information. However, adults' integration of sensory information comes at a cost of losing access to the individual estimates that feed into the integrated percept ("sensory fusion"). By contrast, 6-y-olds do not experience fusion, but are able to keep stereo and texture information separate. This ability enables them to outperform adults when discriminating stimuli in which these information sources conflict. Further, unlike adults, 6-y-olds show speed gains consistent with following the fastest-available single cue. Therefore, whereas the mature visual system is optimized for reducing sensory uncertainty, the developing visual system may be optimized for speed and for detecting sensory conflicts. Such conflicts could provide the error signals needed to learn the relationships between sensory information sources and to recalibrate them while the body is growing.

Fig. 1.

Cue integration experiment. (A) Left eye (LE) and right eye (RE) views of planes with disparity and texture (Upper, condition DT) and disparity-only (Lower, D) cues to slant. To view stereoscopically, free fuse by diverging the eyes (slants from disparity may differ from those in the experiment, where these images took up 13° of visual angle). For texture-only (T) trials, tiled planes were viewed monocularly. (B) Mean ± SEM 75% discrimination thresholds by age group (total n = 92). Circles: mean ± SEM ideal observer predictions for DT. DT thresholds significantly lower on two-tailed paired t test than either D or T are shown by * and significantly lower than both D and T by **. (C) Mean ± SEM relative weighting for texture vs. disparity in conflict conditions with low and high slant. Differences significant at the 5% level are shown by *. Green lines: groups showing mature behavior on each measure.

Fig. 2.

Fusion experiment. (A) Schematic view of five experimental conditions, in which observers made same/different slant judgments for pairs of planes, a standard (shown gray) and a comparison (shown white). On “different” trials (shown), the comparison plane's slant was greater (+) or less (−) than the standard plane's slant, based on single or combined texture (T) and disparity (D) cues. On “same” trials, standard and comparison planes had the same slants, signaled by these same cues. Integration predicts improved discrimination given congruent combined cues vs. the best single cue, but reduced discrimination (“fusion”) given incongruent combined cues vs. the worst single cue. (B) Mean d′ ± SEM, measuring discrimination for same vs. different slants by condition in adults (n = 20) and 6-y-olds (n = 20). Combined-cue conditions T+D+ and T+D− are compared with mean ± SEM d' predicted by integration (yellow circles) and probability summation of single cues (white and black circles). The probability summation model using a “liberal” criterion detects a difference if either single cue signals a difference (white circles). The model using a “conservative” criterion detects a difference only if both single cues signal a difference (black circles). See SI Materials and Methods for details.

Fig. 3.

Latency experiment: Mean ± SEM proportions of correct judgments of the greater slant (Left) and mean ± SEM response latencies (Right) for adults (n = 17) and 6-y-olds (n = 15), comparing pairs of planes defined by disparity (D), texture (T), or both (DT). DT mean differs significantly from either D or T (*) or from both D and T (**) on two-tailed paired t test.