Page 11 of 11 3MM3_P1ReportPart2_06

Results

The purpose of this study was to examine if sensory adaptation would be exhibited in participants involved in a blind walking task; by comparing performance in a visual condition to a non-visual condition and performance in early to later trial blocks. If sensory adaptation is present the hypothesis is that the participant will overestimate distances as compared to visual condition or earlier block of trials.

When considering the non visual (prior exposure) against visual (no prior exposure) variables and the distance estimated by the participants we see that the mean of the non visual (M=10.224) condition is slightly lower than the mean of the visual condition (M=10.368) (Figure 1). At the same time the standard deviation of the non visual condition (SD=3.981) is larger than the visual condition (SD=3892).

The mean distance estimate of block 1 (M=10.543), was the largest followed by the means of Block 3 (M= 10.383) and Block 2 (M=9.961) (Figure 2). The standard deviation of Block 2 (SD= 3.362) was smaller than both the standard deviation of Block 1 (SD= 4.058) and Block 3 (SD= 4.111).

The statistical test used to interpret the data was a repeated measures 2x3 Factorial ANOVA. Examining the two conditions of visual (no prior exposure) and non visual (prior exposure) with performance through three different blocks. There was not a significant result when examining the effect of the condition variable (F(1, 95) = 0.063, p>0.05). The effect of previous trials, the blocks, was also non-significant (F(2,31) = 0.367, p>0.05). Furthermore the interaction between condition and block resulted in yet another non-significant result ( F(2,32) = 0.102, p>0.05).

By further examining the error percentage for each distance between visual non-visual conditions the figure seems to suggest a non-significant result (Figure 3). Also by examining error percentage between Block and Distance it is clear that there is not a significant result (Figure 4).

Discussion

The intent of this study was to shed light on the topic of sensory adaptation and its effect, if any, on distance estimation. The blind walking task was employed as an effective way to study distance estimation, and is a task, which could be influenced by sensory adaptation. The hypothesis which was proposed earlier, is that, if sensory adaptation is present, the participants will overestimate distance, in relation to their estimates previous to sensory adaptation. The experimental design consisted of performance comparisons on 2 different levels, one being the visual condition (control) versus the non visual condition (experiment), and the second being performance comparison during the first, second and third block of trials.

The statistical analysis, which was conducted, did not result in any significant results. The condition variable, using Factorial ANOVA, resulted in a non-significant result, as did the block variable. Furthermore, the interaction between condition and block also proved to be non significant.

The lack of a significant result could be attributed to several flaws and factors in the experimental design. The fact that the data was recorded by four different groups of experimenters could have compromised the data. Even though, all four groups were given the same instructions, many factors may have effected data gathering differently between the groups, such as the level of communication between group members, the exact location of the trials and the confidence and demeanour of the experimenters. All of these factors could have lead to inaccurate results. It was also revealed that Group 4 was forced to use a different subject for day 2 as one of their subjects was hurt, although this may not have the greatest impact on the results it still it is unquestionable that this occurrence was not inline with the experimental design.

The apparatus used could have also lead to the lack of significant results. The blindfold itself had limitations. Firstly, if a participant chose to look down towards the ground he or she could clearly see the ground, and their own feet, it is quite possible that some subjects may have abused this. Secondly, even if they did not abuse this flaw, the blindfold itself allowed some light to pass through, and if the eyes were not closed completely, it may have led to skewed performance as the light striking the eye changed while the participant walked. This change of lighting on the eye, could possibly have an effect on performance, and may have functioned as a primitive cue.

Possibly the largest contributor for the lack of significant results may be the fact that the participants were not naive to the purpose of the experiment. Although, they did not know the exact distances and in which order they would be walking them, the participants did have just as much knowledge and expectations about the study as did the experimenters. This compromises the scientific method and may have greatly impacted the result. It would have been very difficult for the participants to leave their expectations and prior knowledge behind, and not be influenced by it; this in the end may be the most important finding of this study.

Despite these limitations the current study, it is not without its merits. The blind walking task has been a staple of non-verbal distance estimation (Elliott, 1987; Thomson 1983), and has always been seen as accurate and reliable. The effect, which sensory adaptation may have on it, needs to be examined, and was the aim of the current study. Sensory adaptation is a fundamental psychological phenomenon, and saying this there is no reason to believe that it should not be present and have an affect on distance estimation through the blind walking task. The lack of a significant result should be attributed to the experimental and methodological flaws reviewed above and not to the lack of effect of sensory adaptation on distance estimation. The impact of a significant result on the study of distance estimation would be tremendous and this coupled with the fact that a significant result seems to be logical inference, means that changes should be implemented on the experimental and methodological design, which should lead to such a result.

By keeping the experimenters constant amongst all subjects, the effect of individual experimenter difference will be eliminated. The results will therefore become more homogenous and should lead to less experimental error. A better blindfold also seems to be a logical change, which should also lead to more accurate data, it is undeniable that the blindfold was flawed and this may have in turn affected the data. Most importantly the subjects themselves need to be naïve to the purpose of the study and blind to methodology and theory behind the experiment. Only in this way can it be ensured that the data will not be skewed by prior knowledge of the participants. A larger sample size may also be affective in empowering the study and leading to a significant result. If these changes within the experimental and methodological design are enacted, the study holds a much better chance at unravelling a significant result.

References

Elliott, D. (1987). The influence of walking speed and prior practice on locomotor distance estimation, Journal of Motor Behavior, 19, 476-485

Proffitt, D. R., Stefanucci, J., Banton, T., & Epstein, W. (2003). The role of effort in perceiving distance. Psychological Science, 14, 106-112.

Thomson, J.A. (1983). Is continuous visual monitoring necessary in visually guided locomotion? Journal of Experimental Psychology: Human Perception and Performance, 9, 427-443.

Figure 3

Figure 4

Figure 1: Mean and standard deviation of Non visual and visual conditions

Figure 2: Mean and standard deviation of Block 1, Block 2 and Block 3

Figure 3: Average percent of error for each distance for the non visual and visual conditions. Graph suggests non-significant result.

Figure 4: Average percent of error for each distance for Block 1, Block 2, and Block 3. Graph suggests non-significant result.