HOW SOUND AFFECTS THE SHORT-TERM MEMORY AND PERFORMANCE LEVELS: A Quantitative Study

Introduction

This experimental study, conducted by a team of five researchers, was part of our coursework for the “Studying Human Performance” module. It explores the relationship between speech, the acoustics of speech, and short-term memory.

Short-term memory serves as a temporary storage system for information that lasts only for a brief duration. To illustrate this concept, think of the familiar experience of mentally repeating an OTP received via SMS or email until it is typed into the designated field. Almost instantaneously, the OTP vanishes from memory once it has served its purpose. This stands in stark contrast to the ability to remember phone numbers of loved ones, as they both consist of numerical sequences.

The aim of this experimental study was to investigate the capacity of human short-term memory to store random numbers and examine the impact of distracting noise on this ability. Specifically, we conducted a comparative analysis of human performance under three conditions:

a) No distraction during the task,

b) Distraction caused by speech, and

c) Distraction caused by speech-like acoustics (i.e., noise resembling the tonal qualities of speech without the use of actual words).

Methodology

In this experimental study, a total of three conditions were employed to investigate the impact of different distractions on short-term memory. Participants were exposed to three different videos corresponding to each of the three conditions in question.

In each video, a sequence of eight random numbers was sequentially presented to the participants, which they were required to memorize. Following the presentation of the numbers, a period of silence lasting 10 seconds was observed in the no-distraction condition. For the speech distraction condition, participants were presented with a speech on a random topic during the 10-second interval. In the acoustics distraction condition, participants listened to the speech playing in backward, so that they depict only the acoustics of speech. Each condition was repeated ten times to ensure the validity and reliability of the results. Subsequently, participants recorded their answers by writing down the memorized numbers in the correct order.

Scores were then calculated for each experiment, with the score representing the number of correctly recalled numbers out of the total eight presented. By comparing the scores across the different conditions, the study aimed to identify the condition that had the most significant impact on short-term memory performance.

Data Analysis

A total of 40 participants, including friends, family, and classmates, were recruited to participate in this study. Each participant was provided with an Excel sheet to record their answers, with a predefined cell at the end of each row for calculating their score out of 8. However, upon analysis, it was discovered that four participants had obtained a perfect score, indicating potential cheating. Consequently, these participants were identified as outliers and were removed from the dataset. The subsequent data analysis proceeded with a reduced sample size of 36 participants.

Before analyzing whether speech or acoustics of speech affects memory the most, there is a need to ensure whether irrelevant sounds or distractions have an influence over short-term memory. Therefore, we split our problem into two research questions.

Research Question 1:

Does irrelevant sound affect memory?

H0 : Irrelevant sounds does not affect memory

H1: Irrelevant sounds affect memory

The analysis involved comparing the memory performance scores obtained under different conditions (no distraction, speech distraction, and acoustics distraction) using Friedman’s Test.

Research Question 2:

is it the content of the speech that affects memory or is it the acoustics of speech that really matter?

H0: Content of the speech does not affect memory

H1: Content of speech affects memory

The analysis aimed to determine whether there was a significant difference in memory performance scores when participants were exposed to speech distraction compared to acoustics distraction using Wilcoxon Signed Ranks Test

Results

To ensure the data collected is normally distributed, we conducted the Shapiro-Wilk normality test on the collected data. The Shapiro-Wilk test is a suitable statistical test for assessing the normality of data, particularly when the sample size is small.

A graph is said to be normally distributed when the data follows a bell-shaped curve, with the majority of observations clustering around the mean and equal frequencies of observations occurring above and below the mean.

From the graphs plotted above from the data collected, it is evident that only the data collected for Speech condition is normally distributed. The Shapiro-Wilk test also gave us the same result.

From the above table, we can see that the p-value of speech is greater than 0.05 (p>0.05) and the p-value of silence and backward (acoustics of speech) is less than 0.05 (p<0.05) indicating that only speech is normally distributed. Therefore, our data is found to be non-parametric i.e, not normally distributed.

So, we chose appropriate non-parametric statistical approaches for our research questions — a) Friedman’s Test and b) Wilcoxon Signed Ranks Test.

a) Friedman’s Test

Does irrelevant sound affect memory?

H0 : Irrelevant sounds does not affect memory.

H1: Irrelevant sounds affect memory.

For this research question, we considered three levels of the same Independent variable — 1) Silence 2) Speech 3) Backward speech (Acoustics) and here are the test results:

From the above table, we can see that p-value is 0.22 (p<0.050) and we can reject the null hypothesis H0. Therefore, we can conclude that irrelevant sound has an influence on short-term memory. i.e., the influence of silence, speech and acoustics of speech affect the performance on different levels.

b) Wilcoxon Signed Ranks Test

is it the content of the speech that affects memory or is it the acoustics of speech that really matter?

H0: Content of the speech does not affect memory.

H1: Content of speech affects memory.

For this research question, we considered two levels of independent variables, speech and acoustics of speech, and determined which had the worse effect on memory.

From the above table, we can see that the significant value is 0.314 (p>0.05) and we can conclude that there is not enough evidence to reject the null hypothesis. So, the content of the speech does not have a significance impact on short-term memory when compared to the acoustics of the speech.

Conclusion

The findings of the study and their analysis demonstrate unequivocally that irrelevant sounds like speech and acoustics have a significant impact on short-term memory when compared to silence.

Furthermore, any distraction will affect the memory, whether the distraction is meaningful such as speech, or meaningless such as mere acoustics.

Scientific Poster

Here is the scientific poster we submitted as part of this coursework.