Does Depression Impact Our Memory for Rewarding Experiences

Sleep, at its basis, is known to be an important factor for mental health outcomes, yet the contribution of sleep dependent memory consolidation to those outcomes is poorly understood. In depression, learning and memory consolidation is often impaired and this may have long-term consequences that are driven by poor sleep, for those diagnosed with a depressive disorder. Indeed, depressed patients also have impairments in sleep specific physiological activity, which are crucial for efficient memory consolidation whereby memories are reactivated for long-term storage (e.g., slow-wave-sleep). This may contribute to depression symptoms such as anhedonia, a loss of interest in pleasurable activities, which in one study has been related to reduced ability to remember rewarding experiences and reduced activity in reward-related brain networks (e.g., ventral striatum). In order to understand the complex relationship between, sleep, depression and reward it is first necessary to establish a reliable effect size for the relationship between depression and reward-related memory consolidation. Indeed, in recent years it has been found that many published studies lack statistical power to reliably establish robust effect sizes due to small sample size. Therefore, in this experiment the investigators will examine the impact of depression on reward related memory consolidation with a large sample to yield a reliable estimate of the effect size. All participants will undergo the motivated learning task and will have their memory tested for reward related images before and after a 24-hour retention interval. Based on previous research the investigators predict the following:

H1: Memory for images associated with high reward will be greater compared to memory for images associated with low reward.

H3: Healthy controls will have higher memory performance compared to depressed patients.

H2: The magnitude of the benefit for high reward on memory will be greater for healthy controls compared to depressed patients.

Sample Size:

The investigators determined the sample size by performing a data simulation. They achieved this by simulating effects which are in line with our hypotheses using a linear mixed effects model. That data simulation indicated that 400 participants in total would be required to detect the effects of interest with 95% power. If participants have missing data, then that data will be recollected until the desired sample is achieved.

Procedure

All participants will complete the following procedure online via their choice of browser and computer over two sessions. In the first session participants will first sign the consent form then complete the Becks Depression Inventory (BDI) to determine their mental health status. Once complete participants will provide their demographic information (age, sex, gender, education, etc.), complete the Alcohol Use Disorders Identification Test (AUDIT), the Generalised Anxiety Disorder - 7 (GAD-7), Snaith-Hamilton Pleasure Scale (SHAPS), Stanford Sleepiness Scale (SSS) and lastly the Psychomotor Vigilance Task (PVT). Then participants begin the first part of the Motivated Learning Task, the study phase. After a 24-hour retention interval, participants complete the SSS and PVT a second time then complete the test phase of the Motivated Learning Task. Following the conclusion of the test phase, participants are debriefed.

Analysis Plan:

The investigators will use a linear mixed effects models with maximal effects structures to answer the hypotheses. In this case the maximal structure will be used as not including all slopes, intercepts, and correlation structures can increase false positive rates. However, it is also possible that such a model may not converge due to this. Therefore, if necessary, parameters will be removed iteratively starting with correlations of random slopes and then the random slopes themselves.

To examine the hypotheses (H1, H2 and H3) the following linear mixed effects model will be used and reduced accordingly for the reasons specified above:

d' ~ reward * depression + (reward * depression| participant)

Although there will be a positive effect of reward on memory sensitivity in both depressed participants and healthy controls, the investigators expect that this effect will be more pronounced in the latter.