Just noticeable differences in affect: Estimating the minimum change in positive and negative affect that meaningfully impact people’s subjective experience of emotions

Project Information

MEANINGFEEL

Grant agreement ID: 883785

DOI

10.3030/883785

Project closed

EC signature date 2 March 2020

Start date 15 February 2021

End date 16 August 2023

Funded under

• EXCELLENT SCIENCE - Marie Skłodowska-Curie Actions

Total cost

€ 174 806,41

EU contribution

€ 174 806,40

Coordinated by

TECHNISCHE UNIVERSITAET DRESDEN
Germany

Periodic Reporting for period 1 - MEANINGFEEL (Just noticeable differences in affect: Estimating the minimum change in positive and negative affect that meaningfully impact people’s subjective experience of emotions)

Reporting period: 2021-02-15 to 2023-02-14

Summary of the context and overall objectives of the project

Positive and negative affect are two components of experienced emotions that are the foundation of our subjective wellbeing, with important influences on mental and physical health. Self-report measures have been developed to quantify positive and negative affect, which can then be statistically compared within or across groups. They are used as indicators of subjective well-being in large surveys of European populations and as outcome measures in thousands of research studies. However, our understanding of self-reports of affect is limited in several respects. First, we can establish whether there has been a statistically significant change in affect over time, or whether there is a statistically significant difference between people, but, using the results of surveys and research to shape social policies or medical treatments requires the answer to a crucial yet unanswered question: How much do measurements of change in affect reflect people’s meaningful experiences? That is, how much change in self-reported affect reflects a meaningful change to people? Second, there are various biases in self-reports of affect that are little understood. Our limited understanding of self-reports of affect thus limit our ability to interpret research results as well as the accuracy of these results. To address these limitations, my project achieved two main objectives: (1) develop a better understanding of self-report measures of affect, with a focus on estimating the smallest subjectively experience difference in affect that is considered meaningful to people and examining several biases in these self-reports, and (2) examine cross-cultural differences in such bias.

Work performed from the beginning of the project to the end of the period covered by the report and main results achieved so far

I conducted several studies to achieve the objectives. One study was a large, intensive, longitudinal study measuring affect several times per day for a period of 1 week, plus an exit survey asking people about their affect over that week-long period. This study examined 3 things. First, I examined a bias called the memory-experience gap in affect: people's reports for how they felt over the past week tend to be higher, on average, than the average of their affect reports given throughout that week. This memory-experience gap is typically assumed to be due to memory biases. However, at the end of the week, I also asked people to report their affect for each day of that past week. The results were as follows. People's reported affect for the past week was higher than their reported affect for the past day averaged across each day of that same week (i.e. there was a memory-experience gap). However, people's reported affect for each day of the week, given at the end of the week purely from memory, was lower (not higher) than their reported affect for the past day given at the end of each day. Therefore, the memory bias was such that people's reports of their daily affect given at the end of the week was lower than their daily affect given each day. This means that the memory-experience gap is not due to memory biases. Second, I examined a bias in response styles: people's repeated self-reports are biased such that some people give more variable responses than others and this is unrelated to the affect being measured. However, I found little support for such bias, showing that repeated self-reports are actually not biased by response styles. The findings for these two biases are currently written up and under review at top journals. Third, I am using these data to determine the smallest subjectively experienced difference in affect.

Another set of studies were longitudinal studies spanning either 1 day or 2 weeks, where affect was measured at timepoint 1 and timepoint 2. However, whereas some people responded to the affect items at both Time 1 and Time 2, others only completed the measures at Time 2. This means that some people reported affect for the first time at Time 2 and others reported affect for the second time at Time 2. These studies found that people reporting affect for the first time had higher ratings than people reporting for the second time. This means that people's self-report ratings of affect are biased upwards. These findings have been published in a top social psychology journal.

In addition, the results of the studies have been presented at conferences as posters, and to specialist research groups around Europe and in Australia in the form of oral presentations.

Progress beyond the state of the art and expected potential impact (including the socio-economic impact and the wider societal implications of the project so far)

My findings show that self-reports (especially of negative subjective experiences such as) of affect are biased upwards. This impacts findings of research which uses such self-reports and has implications for all such research. This goes beyond the state of the art because, until now, it was believed that this bias was not robust (i.e. not a real bias). Moreover, the methods I developed to test this bias are innovative and may be used by other researchers to examine the phenomenon.

My findings showing that the memory-experience gap is actually not due to memory-biases goes beyond the state of the art because it falsifies an assumption long held by some researchers. For example, some researchers believed that people may overestimate their affective experiences in memory. This theory has been used to draw conclusions about different memory processes and biases in people with mental health problems as compared to people with no such diagnoses. My findings suggest that all such conclusions should be revisited.

Poster