Affective Attentional Bias Training In Depression

The existence of a negative affective bias in depression has been consistently supported throughout literature and has been identified as one of many factors that are associated with depressive symptomatology. A negative affective bias in depression refers to the fact that individuals with depression or depressive symptoms are more likely to focus on negative stimuli and less likely to pay attention to positive stimuli. Typically, this bias is most prominent and consistent in relation to sad stimuli, as opposed to other negatively valenced stimuli, such as anger or fear. Affective bias has been found to play a role in both the onset and maintenance of depressive disorders, and extends to information interpretation, memory, and attention.

Such findings regarding the influence of this negative affective bias on depressive symptoms has spurred further research that aims to reduce or 'correct' this bias in an attempt to improve mood state and reduce depressive symptoms. This research has generally focused on the use of dot-probe attention bias modification (ABM) paradigms, which are forms of cognitive bias modification that attempt to 'train' attention towards positive stimuli and away from negative stimuli through a manipulation of the dot-probe. ABM paradigms have found success in a number of studies, and have been shown to be effective in decreasing the attentional bias for negative information, increasing attention for positive stimuli, and decreasing depressive symptoms leading to improvements in mood. Thus, ABM research represents a promising avenue for treating depressive symptoms through the elimination of the negative affective bias.

However, the diverse methodology, stimuli, and design of existing research has led to conflicting and inconsistent results, with some studies being either unable to manipulate attention at all or failing to significantly improve mood. The use of the dot-probe paradigm as a means of modifying or training behaviour has been contested, with some findings suggesting that it is unable to induce either a positive or a negative attention bias. Moreover, single-session dot-probe paradigms in particular often fail to achieve any significant effects, which may be a result of insufficient durations to produce a change in attention. Additionally, designs that use negatively valenced emotions other than sadness may be weakened due to the fact that an attentional bias towards negative emotions other than sadness in depression has not been consistently demonstrated. Finally, it is common for similar studies to make use of either dysphoric populations or undergraduate populations with varying degrees of non-clinical depressive symptoms. Oftentimes, such studies fail to produce robust effects, which may be a result of an affective bias being less pronounced in non-clinical populations, or a result of a floor effect, in which less-depressed participants with lower initial depressive scores do not experience statistically significant changes. It would be premature to dismiss the effectiveness of attention modification training for affective biases in depression without addressing, clarifying, and improving upon such potential weaknesses in design and implementation.

The proposed novel attention bias modification study seeks to administer a novel training paradigm that utilizes a reward-based, naturalistic, free-viewing design. It is expected that the design of this study will improve upon past research in four ways: 1) by overcoming the limits of the dot-probe through a free-viewing paradigm; 2) by utilizing only stimuli that represent 'sadness' and no other negatively valenced stimuli; 3) by providing multiple training sessions to induce a meaningful effect on attention; and 4) by providing treatment to individuals with MDD who might be more likely to benefit from ABM for a negative affective bias.

This study will take place over five sessions over approximately two weeks: a pre-training assessment, three sessions of affective bias or sham training, and a post-training assessment. Participants will be asked to complete a brief, at-home measure of mood state on each day between pre- and post-assessments. Participants will be told that the study involves attention and learning, but to preserve blinding, they will only be told that the study is specifically an "attention training" study.

Upon arriving in the lab for the pre-treatment session, participants will be asked to read letter of information regarding the study. Once they have read the letter and their informed consent is provided, a trained researcher will collect demographic information and then provide the participant with the BDI, Rumination Scale, and Sheehan Disablity Scale forms to complete privately. Following completion of the forms, participants will be led into a separate room containing the eye-tracker to complete a baseline assessment of affective bias and cognition.

Cognitive assessment: Participants will complete the Emotional Verbal Learning Test . The EVLT is a measure of emotional learning and memory. It involves the oral presentation of emotional words by an experimenter to a participant. Participants will then engage in a short-delay (immediate) and long-delay (after 20 minutes) free recall of the words. Participants will then complete digit span and digit symbol coding tasks, which are widely used measures of attention and will take approximately 7 minutes to complete.

Affective bias assessment: First, affective bias will be assessed using a free-viewing eye-tracking paradigm that is similar in design to the subsequent training sessions. During each trial, a black fixation cross will appear in the centre of the white screen for 500ms. Four unique faces in a drawn from the NimStim Set of Facial Expressions will then appear in a 2x2 grid in the four quadrants of the screen, depicting one positive (happy) face and three negative (sad) faces. The location of the happy face will be randomized to any of the four quadrants over each trial. Participants will be told to freely view the faces and data will be collected via the eye-tracker on variables such as location of first fixation, total fixation duration on each type of valenced image, and total number of fixations on each type of image. Each set of faces will remain on the screen for a total of 3500ms before a new trial will begin. There will be a total of 120 trials, separated into two blocks of 60 trials with a 30-second break between blocks.

Second, the parallel line task will be used as a measure of affective bias towards sad stimuli. In this task, 32 sad stimuli and 32 neutral stimuli will be presented. On either side of the picture, short grey lines will be presented, and each line will be tilted slightly to the left or right. In half of the trials, the lines will be tilted in the same direction (parallel), and in the remaining trials the lines will be tilted in opposing directions. Subjects will be asked either to indicate whether the type of picture being shown is sad or neutral (and thus ignore the lines), or to indicate whether the lines are parallel or different (and thus ignore the picture content). The reaction times to each type of display will be analyzed. Participants will complete 256 different trials for a total task time of about 22 minutes.

Training Sessions: Affective attentional training will be completed by participants in-lab using an eye-tracker in three different sessions over the course of approximately one week. The attention training paradigm to be used was developed for the present study. The modified ABM task was developed to train participants' attention away from negative stimuli and towards positive stimuli through punishment and reward, respectively.

The active training task will include one positively and three negatively valenced faces (showing either happiness or sadness, respectively), selected from the NimStim Set of Facial Expressions, will be presented in a random orientation in a 2x2 grid on the screen. Participants will be instructed to view the images naturally. All four faces will originally appear slightly blurred. Participants will be told that the purpose of the training is to enhance their ability to focus on images that become clearer while avoiding those that become blurrier.

Each trial in the active training group will begin with a black fixation cross in the centre of a white background. The cross will appear alone for 500ms. After the 500ms, the four valenced faces will appear, with one in each quadrant of the screen, equally spaced from the central fixation cross, and participants will begin freely viewing the stimuli. If participants fixate on any of the three negative faces for more than 300ms, the face will undergo a processing of blurring for 500ms or until the participant looks away. This is meant to "punish" the participant by removing the social stimuli, and influence them to adjust their focus to more rewarding stimuli. If the participant fixates on the positive face for 300ms, then this image become clearer over a span of 500 to 1500 ms or until the participant looks away. This is meant to "reward" participants for focusing on positive stimuli. If participants maintain their attention on the positive face until it becomes fully clear, the trial will be successful and the images will disappear, and the program will move on to the next trial. If a positive fixation does not occur within 4500ms, the trial will automatically end and the next trial will begin.

In the sham training group, the paradigm will function exactly the same as the active training group, except that the stimuli will be neutral, non-affective, non-social photos (i.e., cars), with a photo type (i.e., forward facing car) being chosen as the "rewarding" stimulus and the remaining three photos acting as "punishing" stimuli.

Training will be delivered through five five-minute blocks separate by 90-second rest periods. Participants will complete as many trials as they are able within each five-minute block. Each training session should take approximately 35 minutes, for a total of 105 minutes of attention training throughout the course of the week long training.

Post-assessment: The post-assessment will be conducted in-lab during the week following the training sessions. It will consist of a re-administration of the same procedures at baseline.