the effect of walking on the detection of visual stimuli
Movement is life. A hackneyed phrase, which nevertheless very directly points to the fact that a person is constantly going somewhere or coming from somewhere. Despite the fact that walking is commonplace from a physiological point of view, its relationship with perceptual and cognitive functions remains poorly understood. Recent research has shown that although walking appears smooth and continuous, there are critical phases within each step for the successful coordination of perceptual and motor functions. Scientists from the University of Sydney (Australia) conducted a study in which they tried to understand the connection between walking and human visual function. What did the research results show, and where can they be applied in practice? We will find answers to these questions in the scientists' report.
Basis of the study
The world that surrounds a person is not static. Therefore, to survive in a dynamic environment, a person must correctly perceive its changes and respond accordingly to them. At the same time, a person is also dynamic: in one day we can make more than 150,000 saccades (rapid eye movements), 15,000 head turns and several thousand steps.
Despite the thousands of steps we take every day, the impact of movement on perception is largely unexplored. Studies examining the effects of walking and exercise on cognitive performance typically focus on changes in performance over a long period of activity rather than effects occurring within a single stride cycle. These studies showed that when comparing mild exercise to steady state in humans, periods of moderate exercise can produce small increases in performance on cognitive tasks and neuroplasticity. More recent work has shown that walking or light exercise can improve visual processing compared to standing still.
There are many studies that suggest that visual performance may vary across the stride cycle. Recent studies have shown that the response to visual information depends on the phase of human movement. Humans benefit from the pendulum momentum of each step when planning subsequent steps, and visual information about the upcoming area of the future step must be obtained within a critical window before heel contact and that surface for smooth locomotion to occur. In addition, eye movements and the accuracy of hand-eye coordination also show a relationship with the phase of locomotion. Along with each step comes a range of dynamic demands that may lead to modulation of visual processing during walking.
In the paper we're reviewing today, the researchers describe a method that can accurately assess performance on a simple visual detection task during walking to test for modulations in step cycle time. During natural, unobstructed walking, mean accuracy, reaction time, and probability of voluntary responses on a visual detection task were found to fluctuate within the stride cycle predominantly in two frequency ranges (~2 and 4 Hz). These results indicate that constant walking loads induce rhythmic changes in both sensory processing and manual response to each step.
Research results
Image #1
During the experiments, the scientists tested performance on a simple visual detection task while walking naturally in a virtual reality environment. Participants in the experiment had to respond as quickly and accurately as possible to marks appearing in a white ring that drifted slowly and chaotically across the screen.
Throughout the experiment, a continuous adaptive staircase procedure was used that manipulated the target contrast to maintain participants' accuracy at approximately 75%. Confirming this, the average accuracy was close to 75% and did not differ significantly between the participant's stationary position and walking. However, the stationary condition was clearly easier for participants than the walking condition.
Image #2
When contrast was titrated to maintain overall accuracy at 75%, the target contrast in individual trials was presented at ±3 intensity levels centered on the score required for 75% performance, allowing psychometric functions to be fitted across 7 contrast levels. Psychometric fits of participants' accuracy indicated that the difference in accuracy between the stationary and walking conditions was driven by changes in threshold. Interestingly, target reaction time was faster on average during the walking condition compared to the stationary condition. There was also an inverse relationship between reaction time and accuracy at different intensity levels. This indicates that participants were more accurate when responding quickly.
The scientists then combined hundreds of detection data points into a single step cycle to test temporal modulations of visual function. For this analysis, the scientists first broke all walking trials into step cycles (two consecutive steps) by running a peak-detection algorithm on the head height time series recorded in each trial. This analysis showed that the duration of small and large steps varied across participants.
Image #3
The researchers then resampled all steps up to 1–100% step completion to facilitate averaging across participants to assess the main effect of interest—whether the relative phase of an individual's step cycle would modulate visual detection performance. Thus, a target onset that was presented at a random time could be assigned to a position relative to the step at which it occurred (3a–3c). Essentially, hundreds of data points are combined into a single densely sampled step that can be divided into small time bins (such as the 2.5% bin width in the graphs above) and analyzed for temporal modulations.
Step cycle analysis revealed clear fluctuations in performance on the visual detection task. On 3d–3f shows group average results (N = 36) when grouped into 40 linearly spaced, non-overlapping bins. The solid curves represent the best-fitting first-order Fourier model. On 3g correspondence shown (R2) for all Fourier frequencies in the range from 0.2 to 10 cycles per step (cps from cycles per stride; in steps of 0.2 cps). In terms of accuracy, the best-fit performance during the step cycle was 1.93 cps (R2 = 0.48), and group-level agreement was above the 95th percentile of the R distribution2, obtained by permutation (n = 1000) for the range 1.80–2.20 cps. For reaction time, the most appropriate variation was 2.02 cps, and for manual response probability, the most appropriate variation was 1.90 cps.
Image #4
In addition to the group test of stride cycle-based performance fluctuations, the researchers examined the prevalence of stride cycle fluctuations in the sample. Qualitative assessments of the data indicated that some participant-level Fourier fits occurred outside the best group-level fit of approximately 2 cps. Therefore, the scientists quantified the prevalence of significant fluctuations in the range of 0.2-10 cps for each individual participant.
Consistent with group-level results, oscillations at 2 cps were found to be most common in the sample. For example, regarding accuracy fluctuations, 12 of 36 participants had strong matches at approximately 2 cps. Similar results were obtained for reaction time (1.5–2.5 cps) and reaction probability (1.5–2.5 cps).
Few participants showed significant variation across multiple frequency bands on any one measure. Most showed similar individual variation across all three measures, with few showing different variation across performance measures (n = 3/36). On 4a–4f a summary of these data is presented, showing the strength of an individual participant's Fourier transform at each frequency across step cycles. A small group of participants did not show significant fluctuations on each indicator (4g–4i).
Additionally, the possibility that individual frequencies were correlated with stride cycle length was assessed. One possibility is that these fluctuations are not driven by the step cycle, but by a more general oscillator such as time, which when resampled over different step lengths could lead to the observed variability in the peak cps value. As an example, a single 4 Hz oscillator could be recorded as a 2 or 4 Hz oscillation in individuals with step lengths of 500 ms and 1000 ms, respectively. The analysis showed that it is the phase of the step that is important, and not its duration. In accordance with the scientists' hypotheses, significant negative correlations were found between the frequency of perceptual oscillations (in Hz) and the duration of the step cycle.
Taken together, these results indicate that significant variation in performance on visual detection tasks occurs for more than 80% of the sample depending on the phase of the individual's step cycle, and with stable idiosyncratic frequencies of performance per participant.
Image #5
The analyzes described above showed that variation in visual detection performance was widespread across the sample and occurred at stable, idiosyncratic frequencies for each participant. The scientists then tested whether the phase of these oscillations would be the same for all participants.
For this analysis, the researchers included only the subset of participants who showed significant fluctuations and focused on those participants whose fluctuations were in the range of 1.5–2.5 cps. The best-fitting Fourier model was calculated for each participant, and the phase was saved for subsequent tests of heterogeneity. The graphs above show the results of the phase and how individual data are tightly clustered within the phase. Optimal accuracy (5a) and reaction time © occur around the swing phase of the stride cycle, while reaction probability (E) is related to the time of heel contact with the surface.
For a more detailed understanding of the nuances of the study, I recommend taking a look at scientists' report.
Epilogue
In the study, scientists tried to establish a connection between human movement and its ability to simultaneously visually detect objects. During the experiments, participants moved in a straight line in a virtual reality environment, and the stimulus appeared in front of them at different frequencies. Participants were required to respond as quickly as possible to the appearance of the stimulus. Analysis of head height data was used to determine the beginning and end of each stride cycle and to determine the timing of the target relative to the start of the stride. By combining multiple trials and dividing the data into small time intervals, the scientists confirmed that target performance was achieved evenly throughout the stride cycle. However, analyzes of accuracy, reaction time, and reaction probability revealed clear variations in all three performance measures that were systematically related to step phase. Further analysis of individual participants confirmed that these group-level performance variations were present in a significant proportion of the sample, predominantly at 2 or 4 cycles per stride (cps). Taken together, these results indicate a systematic modulation of visual performance within the stride cycle.
It is important to note that oscillation is a ubiquitous phenomenon in neuroscience, both in the brain and in behavior. Previous studies have shown that performance on visual detection tasks waxes and wanes over time. These behavioral oscillations are in the theta band or low alpha band (4–10 Hz) and are differentially associated with the redistribution of attention to perception or decision-making processes. Often in these studies of behavioral oscillations, a salient temporal stimulus is used to “reset” the phase of the ongoing oscillations so that they can be examined from a known and repeatable temporal reference point.
One interpretation of the fluctuations observed in perception and attention is that they occur naturally as a kind of reverberation frequency. From this perspective, period length reflects the time required for signals to travel to higher level areas and then return to sensory levels. A similar picture can occur when moving at a lower frequency, determined by the speed of steps. Obviously, locomotion involves feedforward motor signals, but control of locomotion is impossible without neural and mechanical feedback. There are also numerous studies in rodents showing that locomotion increases the gain of sensory cortical responses in general and visual cortical activity in particular.
Thus, although more studies compare stationary conditions with light exercise or continuous walking, the study authors demonstrated that reliable measurements can be made within the stride cycle. This study revealed clear changes in visual performance associated with the phase of the stride cycle. These results open many avenues for research regarding how and where attention is allocated during the step cycle, whether visual modulations occur uniformly across the visual field, and whether performance on auditory or tactile tasks will also be modulated along with human movement.
A little advertising
Thank you for staying with us. Do you like our articles? Want to see more interesting materials? Support us by placing an order or recommending to friends,
cloud VPS for developers from $4.99
,
a unique analogue of entry-level servers that we invented for you:The whole truth about VPS (KVM) E5-2697 v3 (6 Cores) 10GB DDR4 480GB SSD 1Gbps from $19 or how to properly share a server?
(options available with RAID1 and RAID10, up to 24 cores and up to 40GB DDR4).
Dell R730xd is 2 times cheaper in the Maincubes Tier IV data center in Amsterdam? Only here 2 x Intel TetraDeca-Core Xeon 2x E5-2697v3 2.6GHz 14C 64GB DDR4 4x960GB SSD 1Gbps 100 TV from $199 in the Netherlands! Dell R420 – 2x E5-2430 2.2Ghz 6C 128GB DDR3 2x960GB SSD 1Gbps 100TB – from $99! Read about How to build a corporate infrastructure. class using Dell R730xd E5-2650 v4 servers costing 9,000 euros for pennies?
