Interpretation bias modification training for youth loneliness

Loneliness, often defined as a subjective, unwanted feeling which occurs when we are unsatisfied with the quantity and quality of our social relationships (Paplau & Perlman, 1982), is a common experience among young people. Research suggests that adolescents (Qualter et al, 2015) and young adults (Barreto et al, 2021) are particularly vulnerable to experiencing high levels of loneliness. This is concerning, as loneliness has been causally associated with several negative mental and physical health outcomes (Matthews et al, 2019; Rico-Uribe et al, 2018), and high economic costs (Jeffrey et al, 2017).

However, despite the importance of targeting loneliness early to mitigate its negative outcomes, there are few available developmentally tailored interventions targeting loneliness for young people in the general population (Eccles & Qualter, 2021).

Riddleston et al (2023) recognised this gap in the literature and set out to design an intervention specifically for youth loneliness. They identified one theory of loneliness as a promising intervention target; the idea that some individuals are biased to view uncertain social situations as threatening, leading them to avoid socialising. Other people’s negative reactions towards this social withdrawal then reinforce the individual’s cognitive biases and create a harmful cycle of long-term loneliness (Qualter et al, 2015; Spithoven et al, 2017).

Cognitive bias modification for interpretation (CBM-I) is a technique that focuses on challenging these negative interpretation styles and, instead, encourages neutral or positive interpretations. Riddleston et al (2023) co-designed a CBM-I training programme with young people with lived experience of loneliness, which specifically targeted biases in loneliness-relevant situations. After development, they conducted a preliminary test of the feasibility and acceptability of a single session of the CBM-I training, intending to finalise the protocol so that future studies could more rigorously test its efficacy.

Cognitive biases towards threatening social stimuli are causally associated with chronic loneliness.

Methods

The authors first developed, and then conducted a pilot feasibility trial, of a CBM-I training programme for young people. Participants were recruited from UK-based universities. Eligible participants were aged 18-25; currently enrolled as university students; and had experienced loneliness at some point in their lives.

Stage one: development

The researchers worked with young people with lived experience of loneliness to develop materials for a new interpretation bias measure and to design a CBM-I training programme targeting youth loneliness. 14 young people with lived experience of loneliness participated in an online item generation exercise, resulting in 92 items. 51 different participants then subsequently evaluated these items through an online survey, leading to the selection of 12 items for the interpretation bias measure. The remaining items were adapted for CBM-I training.

Stage two: single session training, feasibility trial

A single session of the digital CBM-I training was delivered to young people experiencing high levels of loneliness to assess its feasibility, acceptability, and potential for altering interpretation bias and reducing loneliness. 56 participants were randomly assigned to receive either CBM-I training or an active control condition:

• In the CBM-I training condition, participants completed an online mental imagery exercise, which involved listening to and visually imagining audio clips of ambiguous social situations which were positively resolved, and then being asked to rate the vividness and positivity of the images. Following this, participants completed the online CBM-I training, which involved listening to audio clips of ambiguous social situations, the majority of which were resolved neutrally or positively, with one remaining ambiguous and one being negatively resolved. Participants then answered questions to test their understanding and were given feedback which reinforced the intended direction of the clip, as well as questions about the vividness and positivity of the images they produced.
• The active control group undertook similar mental imagery training but with the ambiguous scenarios remaining unresolved. They subsequently then completed online control training with similar audio clips of loneliness-related ambiguous social situations, but without being reinforced about which direction to interpret these in. Participants then rated the vividness of the images they produced.

An acceptability survey, along with pre- and post-intervention measures of the newly developed interpretation bias questionnaire and loneliness captured using the revised UCLA Loneliness Scale (Hughes et al, 2004) were then used to assess feasibility, acceptability, and potential efficacy.

Results

56 young people completed the online training session (CBM-I n=29, control n=27). The mean age of participants was 21.6 years, the majority were female (83.9%), and approximately 1/3 were White British/White Other (30.4%). There were no significant differences in age, gender, or ethnicity between the CBM-I and control groups.

Feasibility

Out of 116 young adults who completed screening, 72 (62%) met eligibility criteria based on their loneliness scores. Of the 69 young people invited to participate in training, 61 (uptake = 88%) began the training sessions, with 56 individuals (retention = 92%) completing the training and post-training assessments.

Acceptability

Participants in the CBM-I group indicated that they generally found the training acceptable. Mean rating responses ranged from 3.17 to 3.79 on a scale of 1-5, where higher scores indicated greater acceptability. Responses to open-ended questions pointed to several positive aspects of the intervention, including being encouraged to reframe negative social scenarios positively, the engaging role of mental imagery, and the relatability of the training stimuli. However, participants also highlighted areas that might be improved, such as making the task and scenarios less repetitive, explaining the purpose of the training and the feedback in more detail, and taking cultural background into account.

Potential efficacy

The CBM-I group showed a large reduction in negative interpretation bias and a large increase in positive interpretation bias, along with a large decrease in loneliness post-intervention. In contrast, the control group exhibited smaller changes in interpretation bias and experienced a moderate increase in loneliness.

Participants liked being encouraged to reframe negative social scenarios positively, and the relatability of the stimuli in the CBM-I training.

Conclusions

Riddleston et al (2023)’s findings suggest that delivering this online CBM-I training to address loneliness is viable, feasible and acceptable amongst 18-25-year-olds at university. The preliminary investigation also indicates promising effects of the training on reducing loneliness levels and modifying interpretation biases, the target mechanism of action.

This study suggests that CBM-I training for loneliness is feasible and acceptable for university students.

Strengths and limitations

This study had multiple strengths. Notably, it introduces a novel cognitive bias modification intervention, specifically targeting biases in loneliness-relevant situations. It starts to address the gap in the literature concerning interventions for loneliness amongst young people and has the potential to offer new insights and therapeutic avenues for addressing loneliness. Additionally, the study benefits from a co-creation approach to develop age-appropriate, ecologically valid materials to be used in the CBM-I training. This approach encouraged participant engagement and ensured that the materials developed for the intervention were meaningful to the target population, with several participants describing the relatability of the training stimuli as helpful.

It is also important to consider the limitations of the study, some of which the authors acknowledged. Neither the study design nor the analysis were pre-registered, a process which is important for research transparency and reproducibility. The sample was largely restricted to participants of a narrow age range, who were from UK-based universities, were female, and were white. This limited the generalisability of both the CBM-I intervention, and the findings of its feasibility, acceptability, and potential efficacy, to this specific population. Furthermore, although the UCLA Loneliness Scale has good psychometric properties, evidence suggests that there are differences in what the scale measures across different age groups, or a lack of age invariance (Panayiotou et al, 2021). This prompts doubts about the validity of this measure for young people (Fuhrmann, Riddleston and Verity et al, pre-print) and highlights the need for developmentally sensitive measures of loneliness. In addition, the reliance on self-report measures may have introduced biases such as social desirability bias, the phenomenon in which participants respond in a way that they think is most socially acceptable. There was also the potential for demand effects, in which participants become aware of what the researchers want to find, and for transfer effects, where participants learn skills in one task that then impact their performance in later tasks. One method to reduce the influence of biases would have been to conduct a double-blinded study, in which both the participants and the researchers are unaware of which condition participants have been allocated to. The authors did not report whether methodological steps used to minimise bias, such as blinding or allocation concealment, took place. 

There is a need for developmentally sensitive measures of loneliness that are created specifically for, and with, young people.

Implications for practice

This study has important implications for youth loneliness research and interventions. The findings suggest that CBM-I might be effective in altering cognitive biases associated with loneliness and that this could be implemented on a broader scale, given the high uptake and retention rates among participants. This initial evidence could justify clinicians incorporating CBM-I techniques into their work with young people experiencing loneliness, due to the acceptability and minimal risks associated with the training, and the limited effective interventions that are currently available for addressing this significant public health issue. The creation of a CBM-I digital platform could provide an accessible and cost-effective approach to integrating the training into existing NHS mental health services. However, it is important to keep in mind that these are preliminary findings as part of a small-scale feasibility study. Further research, including a randomised control trial, with larger, more representative samples, multiple training sessions, rigorous methodological steps and follow-up assessments is needed to confirm the preliminary findings and explore the long-term effects of CBM-I training on loneliness and interpretation biases. Nonetheless, if found to be effective, CBM-I could serve as an accessible, affordable, and engaging intervention for youth loneliness, mitigating its negative outcomes both at an individual and societal level.

More research is required to investigate CBM-I training for youth loneliness in samples that are larger and more representative.

Statement of interests

Lauren Turner works at the Youth Resilience Unit at Queen Mary University of London with two of the co-authors of the primary paper (LR and JL). She is currently working as a research assistant on the UNITE project which aims to understand the pathways to loneliness amongst socio-economically marginalised young people, and one of the co-authors of the primary paper (JL) is a co-investigator on this project.

Links

Primary paper

Riddleston, L., Bangura, E., Gibson, O., Qualter, P., & Lau, J. Y. F. (2023). Developing an interpretation bias modification training task for alleviating loneliness in young people. Behaviour research and therapy, 168, 104380. https://doi.org/10.1016/j.brat.2023.104380

Other references

Barreto, M., Victor, C., Hammond, C., Eccles, A., Richins, M. T., & Qualter, P. (2021). Loneliness around the world: Age, gender, and cultural differences in loneliness. Personality and Individual Differences, 169, 110066.

Eccles, A. M., & Qualter, P. (2021). Review: Alleviating loneliness in young people – a meta-analysis of interventions. Child and adolescent mental health, 26(1), 17–33. https://doi.org/10.1111/camh.12389

Fuhrmann, D., Riddleston, L., Verity, L., Alam, I., Conway, J., Chavez, L., … Lau, J. (2023). Co-producing a new scale with young people aged 10 – 24 years: A protocol for the development and validation of the Youth Loneliness Scale (YLS). PsyArXiv. https://doi.org/10.31234/osf.io/2g5ba

Hughes, M. E., Waite, L. J., Hawkley, L. C., & Cacioppo, J. T. (2004). A Short Scale for Measuring Loneliness in Large Surveys: Results From Two Population-Based Studies. Research on aging, 26(6), 655–672. https://doi.org/10.1177/0164027504268574

Jeffrey   K, Abdallah   S, Michaelson   J. (2017) The cost of loneliness to UK employers. London, UK: New Economics Foundation. https://neweconomics.org/2017/02/cost-loneliness-uk-employers/2017

Matthews, T., Fisher, H. L., Bryan, B. T., Danese, A., Moffitt, T. E., Qualter, P., … & Arseneault, L. (2022). This is what loneliness looks like: A mixed-methods study of loneliness in adolescence and young adulthood. International journal of behavioral development, 46(1), 18-27.

Panayiotou, M., Badcock, J. C., Lim, M. H., Banissy, M. J., & Qualter, P. (2023). Measuring Loneliness in Different Age Groups: The Measurement Invariance of the UCLA Loneliness Scale. Assessment, 30(5), 1688–1715. https://doi.org/10.1177/10731911221119533

Perlman, D., & Peplau, L. A. (1981). Toward a social psychology of loneliness. In R. Gilmour & S. Duck (Eds.), Personal relationships, 3, 31-56. Academic Press.

Qualter, P., Vanhalst, J., Harris, R., Van Roekel, E., Lodder, G., Bangee, M., Maes, M., & Verhagen, M. (2015). Loneliness across the life span. Perspectives on psychological science : a journal of the Association for Psychological Science, 10(2), 250–264. https://doi.org/10.1177/1745691615568999

Rico-Uribe, L. A., Caballero, F. F., Martín-María, N., Cabello, M., Ayuso-Mateos, J. L., & Miret, M. (2018). Association of loneliness with all-cause mortality: A meta-analysis. PloS one, 13(1), e0190033. https://doi.org/10.1371/journal.pone.0190033

Spithoven, A. W. M., Bijttebier, P., & Goossens, L. (2017). It is all in their mind: A review on information processing bias in lonely individuals. Clinical psychology review, 58, 97–114. https://doi.org/10.1016/j.cpr.2017.10.003