Global-Local Processing Quantitative Lab Report

Abstract

Visual stimuli are perceived at global and local levels, and people tend to view patterns at the broad rather than the more focused levels. The purpose of the lab report was to investigate the processing of both global and local tasks by undertaking 36 trials for each of the two tasks (Local and global) conditions at three levels (congruent, neutral, incongruent). The report was based on the Navon’s (1977) study. A sample of 15 participants all had normal vision or corrected vision.  The findings have indicated that the patterns within a pair differed either at the global or at the local levels. It was also established that global differences were detected more often in comparison to local differences. Also, the reaction time was higher in global levels compared to the local levels, and this supports Navon’s (1977) study finding that global structuring of a visual scene supersedes analysis of local features in most people.

 Global-Local Processing Quantitative Lab Report

The global pattern is usually responded to faster compared to the local patterns. This is because people voluntarily attend to the global pattern without the effects of the local features, and for them to process the local features; they have to be aware of the whole (Navon 1977). The basic assumption is that a person cannot see the forest for the trees.  Thus, there is an incompatibility between perceptions of wholes and perceptions of details. Both global perception and local perception are influenced by some conditions such as feelings and emotions, but in general, people have a tendency to process globally (Zadra & Clore, 2011).  The purpose of this lab report was to assess the processing of both global and local tasks. At the global level, the visual stimuli are perceived as broad, while the more focused one is the local level or the forest. Navon’s (1977) study was applied to measure the reaction time (accuracy) to small and large letters and compare the reaction times to establish global stimuli and local stimuli.

Method

Participants: A total of 15 participants were used in the experiment, whereby 11 were females and 4 were males. The mean age of the participants was 36.2 years, while the age range was between 19 and 54 years. The participants participated in both global and local tasks. The subjects were requested to be as fast as possible and possibly make minimum errors. None of the participants had participated in similar experiments, and they all had normal vision or corrected vision.

Materials/Apparatus: The equipment used had a screen size of 43.2 cm, a resolution of 800 x 600 pixels, with adjustable height and width of stimuli on screen. The dimensions of the global letters were a height of 25mm and a width of 20mm. The dimensions of the local letters had a height of 3mm and a width of 3mm. In addition, the fixation cross size was 1cm x 1cm and it was displayed for 500ms before the onset of each stimulus. Superlab software was used to offer a better way to build experiments, run them on subjects, and collect data.

Design: An experimental quantitative research method was used. The dependent variable was accuracy (% correct), whereas the two independent variables were Task and Congruence. The task had two levels (global and local), and congruence with three levels (congruent, neutral, incongruent). In the experiment, there were 36 trials for each of the two task conditions. Thus, a total of 72 experimental trials were completed by each subject. In the experiments, the subjects were asked to respond to a presented name of a letter while looking at a visual stimulus that comprised of a large character (which is the global level) and a small character (which is the local level).

Stimuli: Stimuli were presented in random order for each level. In addition, the stimuli were presented for 40 minutes each, followed by a visual mask (dots) until participants made their responses. This was followed by an inter-stimulus interval comprising a blank screen for 100ms. The subjects were required to press the ‘z’ or the ‘/’ key to indicate their response. Participants viewed the screen from a distance of approximately 50 cm. The instructions that were given to participants were for both the local task and the global tasks

  Procedure: 

Navon letters are large, single letters (representing the global perceptual level) that are comprised of smaller letters (representing the local perceptual level). The global and local elements were congruent (a large “T” entailed smaller “T’s”), while the incongruent (a large “T” entailed small “S’s”). Additionally, a Navon task represented a single compound Navon letter on each trial, and the subjects were required to identify either the local letters (small letters) or the global letter (large letters) as quickly as possible. The response time (RT) by each subject about detecting a specific target letter appearing at the global versus local level was compared. In addition, the more global and local interference was also measured (the difference in RT from incongruent to congruent trials) were compared for both local and global trials respectively. Both RT and accuracy were recorded. The RT was measured from the commencement of the inquiry stimulus. Additionally, each of the participants was individually run. The participant was dark-adjusted for five min, followed by a practice session whereby the test session and the control session were carried out. Each of the subjects was allocated one minute rest period min after every block of the trials.

Results

The subjects were instructed on the shape of presented patterns and asked to respond “same” only in case of perfect congruence, neutral if no difference, and incongruent in case there was a difference between the two. The means and standard deviations for all conditions (Congruent, Neutral, and Incongruent) are presented in Table 1 below

Table 1: Mean and Standard Deviations for Global and Local Tasks

As shown in Table 1, the mean and standard deviations for all the Global tasks were higher in comparison to the Local task.

Accuracy

The accuracy levels (% correct for the) Global and Local Tasks are presented in Table 2 below based on the number of participants.

 Table 2: Accuracy of the Global and Local Task

Participants in all groups performed the search task but the accuracy varied between the global and local tasks. For instance, for one participant, the accuracy was 72.72% under congruent, compared to 83.33 under the local task. In addition, when the participants were 6, the incongruent was 100% accurate for both global and local tasks. Participants were more accurate at the global (mean = 75.13, SD = 31.56) than the local task (mean = 57.78, SD = 28.59) [F(1,14) = 4.72, p = .048].

 Table 3: Reaction Time

Table 2 shows reaction time scores for each participant, for each condition. The correct responses were included in the average, and since all the cells are filled, it means that the participant got the answers correct in that condition.

Under the TASK, there was a significant effect as the local task was greater than the local task (M 76.107, 2.725 vs. Std 4.291, 2.725 at a 95% confidence interval). The congruence at the three levels had the following means and Standard deviations Congruent (74.9, 3.36), neutral (72.8, 3.513 and incongruent (62.6, 3.94). Thus, the subjects reacted faster to global patterns compared to local patterns. Under the Task vs congruence, the results for global and local trials were compared respectively under the three levels of congruence. For instance, the means and standard errors for the global task were higher compared to the local tasks at all three congruence levels. (Refer to Table 4 below.

Table 4: TASK Vs. Congruence

Discussion

It is widely documented that global/local processing could be altered via the application of task or stimulus manipulations (Chen, Nakayama, Levy, Matthysse, & Holzman, 2003; Kimchi, 1992; Kinchla &Wolfe, 1979). People also tend to restrict their attention to the whole, rather than the local, and it is after the completion of the categorization of the global pattern that they move to the local patterns (Navon, 1997; Sripati, & Olson, 2009). The purpose of the present study was to investigate the processing of both global and local tasks by undertaking 36 trials for each of the two tasks (Local and global) conditions at three levels (congruent, neutral, incongruent).

The findings of this study have been supported by Dale and Arnell (2014) as well as Navon (1997) that there is a difference between local and global stimuli. For instance, people are more likely to see the global, rather than the local patterns. This study has similarly shown that a Navon letter task can influence performance on the task as established by Weston and Perfect (2005) after participants were induced into a globally and locally focused state. Concerning accuracy, there were significant differences in accuracy in both patterns and depended on the number of participants. This finding is supported by Hills and Lewis’s (2009) study that showed that there was a difference in accuracy after local Navon trials were repeated and that shapes with high spatial frequency information were hard to identify. This supports Navon’s (1997) idea that a local processing bias can minimize face recognition accuracy, while on the other hand, a global processing bias promotes face recognition accuracy.

The difference in RT from incongruent to congruent trials was compared between the local and global trials as shown in Table 2 and there was a significant difference in RT. For instance, the RT for global tasks was higher than for local tasks for congruent levels. About the TASK, there was a significant effect as the local task was greater than the local task whereby the congruence at the three levels differed. The comparison between Task vs congruence shows that the global and local trials under the three levels of congruence are different.

In conclusion, patterns within a pair differ at global and the local levels. In addition, global differences tend to be detected more often when compared to local differences.

References

Chen, Y., Nakayama, K., Levy, D., Matthysse, S., Holzman, P. (2003). Processing of global, but not local, motion direction is deficient in schizophrenia. Schizophrenia Research, 61, 215 – 227. doi:10.1016/S0920-9964(02)00222-0.

Dale, G., & Arnell, K. M. (2014). Lost in the forest, stuck in the trees: Dispositional global/local bias is resistant to exposure to high and low spatial frequencies. Plos One, 9(7), 1-10. doi.org/10.1371/journal.pone.0098625.

Hills, P. J, & Lewis, M. B. (2009). A spatial frequency account of the detriment that local processing of Navon letters has on face recognition. J Exp Psychol Hum Percept Perform 35(3), 1427–1442. Available: http://www.ncbi.nlm.nih.gov/pubmed/19803647. 

Kimchi, R. (1992). The primacy of wholistic processing and global/local paradigm: A critical review. Psychology Bulletin 112, 24–38.

Kinchla, R. A. & Wolfe, J. M. (1979). The order of visual processing: “Top-down,” “bottom-up,” or “middle-out. Percept Psychophys 25(2), 225–231.

Navon, D. (1997). Forest before trees: The precedence of global features in visual perception. Cognitive Psychology, 9(3), 353–383. doi.org/10.1016/0010-0285(77)90012-3.

Sripati, A. P., & Olson, C. R. (2009). Representing the forest before the trees: A global advantage effect in monkey inferotemporal cortex. The Journal of Neuroscience : The Official Journal of the Society for Neuroscience, 29(24), 7788–7796. doi.org/10.1523/JNEUROSCI.5766-08.2009.

Weston, N. J., & Perfect, T. J. (2005). Effects of processing bias on the recognition of composite face halves. Psychon Bull Review 12,1038–1042.

Zadra, J. R., & Clore, G. L. (2011). Emotion and perception: The role of affective information. Wiley Interdisciplinary Reviews. Cognitive Science, 2(6), 676–685. doi.org/10.1002/wcs.147.