N1 Adaptation: Exploring the Neuronal Basis of the Interaction Between Auditory Sensory Memory and Attention

N1 Adaptation: Exploring the Neuronal Basis of the Interaction Between Auditory Sensory Memory and Attention

Gengjie Jack Ho

2023

The aim was to explore whether voluntarily focusing on repetitive auditory stimuli influences the lifetime of N1 adaptation, which indexes the lifetime of auditory sensory memory. Twenty-six neurotypical participants with self-reported normal hearing were recruited from Lancaster University. Electroencephalogram (EEG) recording took place in a sound-attenuated laboratory. A two-by-two factorial design was employed, where one factor manipulated the presence or absence of attention, whereas the other factor manipulated the stimulus-onset interval (SOI), which primarily served to calculate the lifetime of adaptation. Three different amplitude measurement methods were used to calculate the N1 amplitude, therefore three sets of statistical analyses were performed for each investigation. For the preliminary investigation, two-way ANOVAs were conducted to evaluate the impact of attentional focus (presence or absence) and SOI (short or long) on the amplitude of N1. For the primary investigation, paired-samples t-tests were conducted to evaluate whether the presence or absence of attention influences the N1 adaptation lifetime. The preliminary results indicated no significant difference in N1 amplitude between the presence and absence of attentional focus. There was also no significant difference in the SOI, except for one of the amplitude measurement methods, which showed greater N1 amplitudes in the Long SOI condition. The primary results indicated that whether attention was present or not showed no significant effect on the adaptation lifetime across all three amplitude measurement methods. However, the study suffered from low statistical power and possible issues with the methodological design due to the combined use of visual and auditory modalities to manipulate attentional focus. Therefore, it is inappropriate to draw conclusions from the findings of this study. Methodological improvements and theoretical implications were discussed.

neuropsychology, attention, auditory sensory memory, N1 adaptation, sensory processing, neural responses

Methods Section:
Participants
Twenty-six neurotypical participants with self-reported normal hearing (9 males, 16 females, 1 prefer not to say), all of whom were students from Lancaster University, were recruited using opportunity sampling via advertising on social media platforms and SONA. The age range of the participants spanned from 18 to 34 years (M = 22.85, SD = 2.55). Sixteen participants were excluded due to excessive electric noise, resulting in a remaining pool of 10 participants. All participants provided written consent and volunteered to participate in the experiment. The study received ethical approval from Lancaster University’s Department of Psychology.
Stimuli
The experiment employed the oddball paradigm to elicit auditory responses. The standards were presented at a constant rate of 210 repetitions per condition, while the deviants appeared unpredictably at a 5% probability (10 deviants per condition). The sequence of standards and deviants remained consistent across all conditions. The standards were presented as a 500-Hz pure tone, while the deviants were a 503-Hz pure tone. The duration of each tone was 100 milliseconds, with 10 milliseconds of linear onset and offset ramps. All tones were presented at a consistent and comfortable volume level (28% volume on Windows 10). The auditory stimuli were programmed and delivered using MATLAB.
Design
The study followed a two-by-two factorial design (see Figure 1). It included two attention conditions: Active and Passive. In the Active condition, participants were presented with a stream of standards and deviants while focusing on a fixation cross. Their objective was to count the occurrences of deviants. In the Passive condition, participants viewed a nature documentary displayed on a smartphone screen. Their objective was to count the number of animal species featured in the documentary while ignoring the stream of auditory stimuli playing simultaneously in the background. Both the fixation cross and the smartphone screen were positioned one metre in front of the participants. Additionally, there were two SOI conditions: Short SOI (1.7 seconds) and Long SOI (3.4 seconds). The oddball paradigm was integrated into a stimulus block design - with two types of stimulus blocks, each having a specific SOI. Note that the order of the conditions was randomized among participants.
The purpose of the design was to manipulate attention towards repetitive auditory stimuli and calculate adaptation lifetime. The counting tasks in the Active and Passive conditions manipulated attentional focus. In the Active condition, the count-the-deviants task aimed to maintain participants’ attention on the repetitive auditory stimuli. In the Passive condition, the count-the-animal-species task aimed to divert participants’ attention away from the repetitive auditory stimuli using visual stimuli in the form of a nature documentary. Additionally, the counting tasks served as a quality control measure, excluding participants whose answer substantially differed from the correct answer. Conversely, the inclusion of both short and long SOI measured adaptation lifetime using the amplitude ratio (explained below in Data Analysis).
Figure 1. A visual representation of the study’s two-by-two factorial design, encompassing four distinct conditions: Active with Short SOI (1.7s), Passive with Short SOI (1.7s), Active with Long SOI (3.4s), and Passive with Long SOI (3.4s).
Procedure
EEG was used as the method of data collection. The Enobio NIC2 suite recorded EEG data, using three dry electrodes (Fpz, Cz, and Fz) to capture neuroelectrical activity in the auditory cortex (Neuroelectrics, n.d.). Data recording was conducted in a sound-attenuated laboratory. The entire experiment lasted approximately 60 minutes, which included a 20-minute preparation period.
Before the experiment, participants were sent an information sheet online and completed a consent form upon arrival. They were then fitted with an electrode cap and headphones, and instructed to avoid excessive movement during recording to minimise muscle artifacts. When recording was ongoing, participants were verbally given instructions at the start of each condition, and they were asked about their answers to the counting tasks after each condition. Short breaks were allowed when transitioning between conditions. After the experiment, participants were inquired about their age and gender, and received a verbal and written debrief regarding the true purpose of the study.
Data Analysis
We conducted a priori power analyses using G*Power 3.1. to determine the required sample size for testing the two hypotheses (Faul et al., 2007). For the preliminary investigation, results indicated that the required sample size to achieve 80% power for detecting a medium effect, at a significance criterion of α = .05 was N = 36 for a two-way ANOVA. For the primary investigation, results indicated that the required sample size to achieve 80% power for detecting a medium effect, at a significance criterion of α = .05 was N = 34 for a paired-sample t-test. Our recruitment target of 36 participants was based on the larger of the two required sample sizes.
In data preprocessing, we discarded the first few trials from each condition to minimise initial variability in orienting and habituation effects, and excluded any unidentifiable N1 responses.
Measuring the N1 amplitude is essential for estimating adaptation lifetime and conducting the planned data analysis. There are three methods available - N1, N1-P2, and mean voltage displacement. Notably, baseline correction was performed as a standard initial procedure, addressing a baseline that extended over 100 milliseconds within this experiment. The first method identifies and measures the N1 amplitude as the point of maximum negativity (Marton et al., 2018). The second method measures the peak-to-peak amplitude difference between N1 and P2, as it captures the relationship between the two and avoids the problem of a noisy baseline by not depending on the pre-stimulus baseline (Al-Abduljawad et al., 2008; Scaife et al., 2006). The third method estimates the mean voltage displacement (absolute amplitude value) over a specific time frame, particularly useful when the N1 component is difficult to identify, or the stimulus onset is ambiguous (Hoehne et al., 2020; Komssi et al., 2004). All three methods were employed to conduct a more comprehensive data analysis, given that consistent findings across different methods increase the reliability of results and inconsistencies can guide further investigation.
In the traditional approach for estimating adaptation lifetime, one uses multiple stimulus blocks, each featuring varying SOIs ranging from 0.5 to 10 seconds. The ERP is derived separately for each stimulus block, and notably, the peak N1 amplitude is plotted as a monotonically increasing function of SOI. This relationship between the N1 amplitude and the SOI can be described as an exponentially saturating function, represented by the model equation A(1-e-(t-to)/τ), where A (amplitude), τ (time constant), and to (time origin) represent fitting parameters (Lü et al., 1992). Graphically, one fits the exponentially saturating curve to the measured N1 amplitudes. Here, the fitting parameter τ characterizes the steepness of the curve in seconds. τ signifies the SOI at which the amplitude curve reaches 66% of its way towards the saturation limit, indicating the lifetime of adaptation. However, this method is time-consuming and difficult for participants, insofar as boredom-induced mind wandering may confound the effects of attentional focus (Eastwood et al., 2012; Meier et al., 2023).
An alternative approach of amplitude ratio only used two stimulus blocks with contrasting SOIs. By graphically plotting the amplitude ratio of a short SOI against a long SOI over a range of τ values (measured in seconds), it shows that the ratio is a monotonically increasing function of τ. Although this ratio-to-τ relationship is not strictly linear, it can be used to estimate the adaptation lifetime rather than the conventional time constant, given that the ratio increases as τ increases. In practical terms, both SOI conditions produced a clear difference in amplitude. The short SOI of 1.7 seconds ensures a distinct ERP with an observable N1 component (if the SOI is less than 300 milliseconds, it would render the N1 response too minute and difficult to observe), while the long SOI of 3.4 seconds brings the N1 amplitude closer to its saturation limit. By shortening the experiment duration, this ‘dimensionless’ measure addressed the limitations of the traditional approach without significantly compromising estimation accuracy.
Two-way ANOVAs were conducted to assess how the N1 amplitude is influenced by attentional focus (presence or absence) on repetitive auditory stimuli and SOIs (short or long).
Paired samples t-tests were conducted to assess if the presence or absence of attentional focus on repetitive auditory stimuli significantly affects adaptation lifetime (calculated via amplitude ratio).

Lancaster University

Data/SPSS.sav

Ho2023

Sharon Boyd

Open

None

English

Data

LA1 4YF

Patrick May

MSc

Neuropsychology

Participants: 26
Excluded Participants: 16
Final Sample: 10 Participants

ANOVA, t-test