Deterministic and Stochastic Postural Processes: Effects of Task, Environment, and Age

ABSTRACT

Upright standing is always environmentally embedded and typically co-occurs with another (suprapostural) activity. In the present study, the authors investigate how these facts affect postural dynamics in an experiment in which younger (M age = 20.23 years, SD = 2.02 years) and older (M age = 75.26 years, SD = 4.87 years) participants performed a task of detecting letters in text or maintaining gaze within a target while standing upright in a structured or nonstructured stationary environment. They extracted the coefficients of drift (indexing attractor strength) and diffusion (indexing noise strength) from the center of pressure (COP) time series in anteroposterior (AP) and mediolateral (ML) axes. COP standard deviation decreased with drift and increased with diffusion. The authors found that structure reduced AP diffusion for both groups and that letter detection reduced younger SD _AP (primarily by diffusion decrease) and increased older SD _ML (primarily by drift decrease). For older and younger participants, ML drift was lower during letter detection. Further, in older letter detection, larger visual contrast sensitivity was associated with larger ML drift and smaller SD _ML, raising the hypotheses that ML sway helps information detection and reflects neurophysiological age.

Keywords:

• ageing
• drift-diffusion analysis
• suprapostural task
• visual factors

ACKNOWLEDGMENT

This research was supported by the Collaboratory for Rehabilitation Research through a grant from the Provost's Office at the University of Connecticut.

Notes

1. In respect to the clinical evaluation of visual detection capability, contrast (specifically Michelson Contrast) is a measure of the difference between the luminance of an object and the luminance of its surround. It is calculated as the luminance difference between adjacent regions divided by the luminance summed across the adjacent regions. This yields a dimensionless number, typically reported as a percentage. Threshold contrast is that amount of contrast where a participant cannot discriminate between adjacent regions, typically between a sine wave and a patch of uniform grey (50% point on the psychometric function). Contrast sensitivity is defined as the inverse of the threshold contrast. Thus, the lower the contrast detectable by a participant over any region of the visual environment, the higher is their contrast sensitivity in that setting. Contrast sensitivity differs as a function of the spatial frequencies available. Spatial frequency is specified in terms of the size of sine wave grating over the back of the eye using cpd of visual angle.

2. Visual acuity (as measured with a standard Snellen chart) is a special case of VCS under high illumination and spatial frequency conditions that are relevant to the detection of refractive errors but less relevant to visually guided tasks in other embedding environments (for an extensive review, see Ginsburg, 2003).

3. Analysis of the voltage output from the dual-force plate system revealed high frequency spikes, most likely due to measurement noise. A triangular filter with a seven-point width was sufficient to eliminate them. Our prior comparisons of filtered and unfiltered postural sway have not revealed effects of the triangular filter on the pattern of results. An exception is applications of recurrence quantification and detrended fluctuation analyses in which triangular filtering has brought patterns in the unfiltered data into sharper relief. A concern with respect to frequency domain filters is that when recovery of temporal structure is the goal, their use to address measurement noise may inject complexities into the time domain.

4. For examples of the quantitative development of drift-diffusion analyses for other purposes, see Friedrich and Peinke (1997), Frank, Friedrich, and Beek (2006), and van Mourick, Daffertshofer, and Beek (2006, 2008).

5. After averaging the four trials in each condition, the box plot showed an outlier for SD _AP in the no-structure–text condition. It resulted from a single trial. In this trial, SD _AP was more than 7.5 times above the average of all the conditions. The outlier disappeared when this trial was removed from the analysis.

6. That two participants with near 20/20 Snellen acuity exhibited zero contrast sensitivity at 18.0 cpd was most probably due to illumination differences across the test sessions. The clinical protocol to assess contrast sensitivity using the VSRC test station calls for an illumination level that is at the low end of that supporting photopic vision (85 lx, per the manufacturer's benchmarking, Ginsberg's research, and the standards accepted by the United States Air Force in the 1980s). The Snellen acuity assessment, when appropriately standardized, is conducted at 400–480 lux.

7. One possibility for the difference is that the older participants in Prado et al.'s (2007) experiment were younger neurophysiologically than were the older participants in the present experiment. That is, their VCS placed them at the higher end of the VCS axis in Figure 5A.

8. Neuropathy means damage to sensory or motor nerves. Neuropathy thus reduces perception and action capabilities.