Computerized CBT and Self-Esteem

Self-Improvement · Andreas Grivas, MSc Clinical Psychologist · April 22, 2023

Computerized CBT and Self-Esteem

Self-esteem refers to the overall judgement we make about our own worth as a person. Everyone has moments of self-doubt, but for some people a harsh, negative view of themselves becomes a lasting lens through which they interpret almost everything: a compliment is dismissed as politeness, a mistake becomes proof of inadequacy, and new challenges are avoided for fear of exposure. Low self-esteem is not a psychiatric diagnosis in itself, but it is a feature of many difficulties, including depression, social anxiety and eating disorders. It may also be a risk factor. A meta-analysis of longitudinal studies, covering 77 studies on depression and 18 on anxiety, found that low self-esteem predicted later depression (β = -0.16) more strongly than depression predicted later self-esteem (β = -0.08), supporting the idea that low self-esteem can make people more vulnerable to depression; for anxiety, the effects in each direction were more balanced (Sowislo & Orth, 2013).

The most widely used cognitive-behavioural account of low self-esteem, developed by Melanie Fennell, describes how early experiences such as criticism, bullying, neglect or a lack of warmth can lead to a deep-seated negative belief about the self, sometimes called the "bottom line": "I am not good enough", "I am unlovable", "I am a failure". To cope with this belief, people develop rules for living, such as "I must never make mistakes" or "If I say what I think, people will reject me". When a situation threatens to break these rules, anxious predictions arise and the person may avoid the situation or rely on safety strategies, such as over-preparing, people-pleasing or staying silent. When things go badly, harsh self-critical thinking confirms the original belief. In this way, the belief is protected from the evidence that might challenge it.

Cognitive-Behavioural Therapy (CBT) for low self-esteem works directly on this cycle. It helps people identify their bottom line and the rules that flow from it, test anxious predictions through behavioural experiments, notice and respond to self-critical thoughts, deliberately record evidence of their positive qualities, and build more balanced rules and a kinder, more realistic view of themselves. A systematic review and meta-analysis of interventions based on Fennell's model identified eight studies, seven of which were included in the quantitative analysis. Weekly group or individual CBT sessions produced a large improvement in self-esteem (effect size 1.12), while one-day workshops produced a smaller one (0.34), with comparable reductions in depressive symptoms (Kolubinski et al., 2018). The authors noted, however, that outcomes relied on self-report and that it is unclear how different these interventions are from CBT for depression.

Because CBT for low self-esteem is structured and skills-based, it lends itself to digital delivery. In a pilot randomised controlled trial, 52 adolescents aged 15 to 19 were assigned to seven weeks of therapist-supported internet-based CBT or to a waiting list; the treatment group reported substantially higher self-esteem afterwards (d = 1.18), along with improvements in self-compassion, quality of life, depression and anxiety (Berg et al., 2022). A subsequent randomised trial with 86 young adults aged 18 to 25 found a similarly large effect on self-esteem (d = 1.24) for seven weeks of therapist-guided internet CBT compared with a waiting list, with moderate to large effects on secondary measures (Berg et al., 2026). Both research teams describe their results as preliminary and in need of replication in larger samples.

Evidence for self-help smartphone apps is more mixed. In a randomised trial with 400 university students reporting depressive symptoms, four weeks of access to a self-help app combining CBT, mindfulness, acceptance and commitment therapy and metacognitive training led to a greater increase in self-esteem (d = 0.40) and a greater reduction in depressive symptoms (d = 0.26) than a waiting list in the per-protocol analysis (Bruhns et al., 2021). However, when the same app was offered for four weeks to 159 adults after discharge from inpatient treatment for depression, there was no significant effect on the primary self-esteem outcome; only an exploratory analysis suggested a small benefit for one aspect, self-competence (d = 0.35). In that study, people with more positive attitudes and expectations towards digital interventions used the app more often (Bruhns et al., 2023). This suggests that low-intensity apps may help some people with milder difficulties, but should not be expected to shift entrenched negative self-beliefs on their own, especially in more severe depression.

What can digital tools realistically add? Change in self-esteem depends on repeated practice in everyday life: catching self-critical thoughts as they happen, keeping a daily log of positive qualities and achievements, planning behavioural experiments and reflecting on what actually happened. Apps and conversational agents can prompt these exercises, offer a structured space to reflect, and make progress visible over time. At the same time, robust trials of AI chatbots designed specifically to improve self-esteem are lacking, and simple positive affirmations without testing beliefs against real experience are not the same as CBT.

Several limitations should be kept in mind. The digital trials so far are relatively small, mostly compare treatment with a waiting list rather than an active alternative, rely on self-report questionnaires and have short follow-up periods. Internet-delivered treatments are not risk-free: a significant minority of people experience negative effects, although deterioration rates appear similar to those in face-to-face therapy (Andersson et al., 2019). Low self-esteem can also be a sign of something that needs more specialised help, such as major depression, social anxiety disorder, an eating disorder, or the after-effects of trauma or abuse. Where self-esteem is being worn down by an ongoing situation, such as bullying or an abusive relationship, the priority is safety and support, not only changing one's thoughts.

Consider speaking to a psychologist or doctor if negative beliefs about yourself are affecting your relationships, studies or work, if you are avoiding many situations, if you notice persistent low mood, changes in eating, or loss of interest, or if a self-help programme has not helped after a fair trial. If self-criticism turns into thoughts of harming yourself or of not wanting to live, contact your local emergency number immediately. Digital tools can complement, but do not replace, professional assessment and therapy.

Low self-esteem is common, painful and closely linked to depression and anxiety, but it is also changeable. CBT based on Fennell's model has encouraging evidence, and early randomised trials suggest that therapist-guided internet CBT can produce meaningful improvements in adolescents and young adults, while results for stand-alone apps are mixed. Used as a structured way to practise new ways of relating to oneself, ideally with some human guidance, digital CBT can be a helpful step towards a more balanced and compassionate view of the self.

References

  • Andersson, G., Titov, N., Dear, B. F., Rozental, A., & Carlbring, P. (2019). Internet-delivered psychological treatments: From innovation to implementation. World Psychiatry, 18(1), 20–28. https://doi.org/10.1002/wps.20610
  • Berg, M., Lindegaard, T., Flygare, A., Sjöbrink, J., Hagvall, L., Palmebäck, S., Klemetz, H., Ludvigsson, M., & Andersson, G. (2022). Internet-based CBT for adolescents with low self-esteem: A pilot randomized controlled trial. Cognitive Behaviour Therapy, 51(5), 388–407. https://doi.org/10.1080/16506073.2022.2060856
  • Berg, M., Lindegaard, T., Käll, J., Granath, V., Fält, K., Karlsson, M., & Andersson, G. (2026). Internet-based CBT for young adults with low self-esteem: A randomized, controlled study. Journal of Behavior Therapy and Experimental Psychiatry, 94, 102140. https://doi.org/10.1016/j.jbtep.2026.102140
  • Bruhns, A., Baumeister, A., Demeroutis, G., Jahn, H., Willenborg, B., Shaffy, A., Moritz, S., & Bücker, L. (2023). A mobile-based aftercare intervention to increase self-esteem in inpatients diagnosed with depression: A randomized controlled trial. Psychotherapy Research, 33(6), 783–802. https://doi.org/10.1080/10503307.2022.2157226
  • Bruhns, A., Lüdtke, T., Moritz, S., & Bücker, L. (2021). A mobile-based intervention to increase self-esteem in students with depressive symptoms: Randomized controlled trial. JMIR mHealth and uHealth, 9(7), e26498. https://doi.org/10.2196/26498
  • Kolubinski, D. C., Frings, D., Nikčević, A. V., Lawrence, J. A., & Spada, M. M. (2018). A systematic review and meta-analysis of CBT interventions based on the Fennell model of low self-esteem. Psychiatry Research, 267, 296–305. https://doi.org/10.1016/j.psychres.2018.06.025
  • Sowislo, J. F., & Orth, U. (2013). Does low self-esteem predict depression and anxiety? A meta-analysis of longitudinal studies. Psychological Bulletin, 139(1), 213–240. https://doi.org/10.1037/a0028931