top of page

USING COMPUTATIONAL PSYCHIATRY TO CHARACTERIZE THE NEUROPSYCHOLOGICAL PATHWAYS OF ADHD

NOVEL TOOLS FOR PERSONALIZING ASSESMENTS AND TREATMENTS

Objectives and approach

The purpose of my research program is to better understand the neuropsychological pathways of child and adult attention-deficit hyperactivity disorder (ADHD) as well as other mental health conditions such as mood disorders. Specifically, I pursue the following three aims:

  1. Characterizing the neurocognitive processes of ADHD by considering different symptom profiles (e.g., inattention, hyperactivity-impulsivity) and co-occurring mental disorders (e.g., depression, anxiety) to which I refer to as “comorbidities”

  2. Predicting the development of ADHD over time

  3. Providing tools to personalize treatment

To pursue these aims, I am using a multidisciplinary approach that consists of clinical questionnaires, behavioral tests, and physiological measures such as electroencephalography (EEG) and eye tracking. To integrate these different measures into one framework, I am using computational modeling and machine learning. In so doing, I am implementing the Research Domain Criteria (RDoC) approach of the National Institute of Mental Health into practice.(1) Figure 1 summarizes the objectives and the methodological approach of my research program.  

 

Picture1.png

Figure 1. Objectives and methodological approach of my research program.  

“Comorbidities”: ADHD and frequently co-occurring disorders such as anxiety and depression.(2-3) “Persistor”: a child, diagnosed with ADHD, who maintains an ADHD diagnosis into adulthood. “Desistor”: a child, diagnosed with ADHD, who sufficiently improves to not maintain an ADHD diagnosis into adulthood. “Controls”: a child without ADHD. 

Significance of my research program

Characterizing ADHD from a multidisciplinary perspective can help in the search for effective, tailored assessments and treatments, especially identification of subgroups responsive to specific treatments.(3) Developing an integrative modeling tool allows psychologically meaningful interpretations of neurocognitive and EEG measures, as well as eye-tracking patterns. Providing this tool also sets the groundwork to eventually incorporate other measures (e.g., genetics, microbiome). My research thus far focused on child as well as adult ADHD. I also put a stronger emphasis on comorbidities because the neurocognitive characteristics of ADHD and comorbidities (particularly adults) remain unclear.(3-5)

​

Components of my research program

My research program involves multiple (ongoing/completed) projects. Each completed project (see my section: publication) helped me to develop the integrative approach outlined in Figure 1. Currently, I am implementing this integrative approach as part of my dissertation.

Dissertation Project

In my review article,(3) I found that young adults with ADHD (particularly females) represent an understudied population. Studying the neurocognitive characteristics of this population is important given that the number of young adults diagnosed with ADHD is increasing.(10-11) For my dissertation, I use a novel test environment (composed of self-developed cognitive and social-cognitive tasks, physiological and clinical measures) to pursue the following aims:

  1. Characterize neurocognitive processes of ADHD comorbidities

  2. Study how individual differences relate to symptom severity

  3. Develop new methods for assigning psychological interpretation to physiological measures

The novel test environment provides a better understanding of the neuropsychological characteristics of adulthood ADHD (including comorbidities). I use a joint-modeling approach (explained below) to link behavioral and physiological measures. Ultimately, I will apply this modeling approach to the data of randomized clinical trials to index different developmental courses of ADHD.

Explanation of methodological approach

Computational modeling

So far, I mostly focused on the diffusion decision model which belongs to the class of sequential sampling models (SSMs). These models can be used to decompose behavioral performance into distinct mental components (Figure 2: z, v, a, Ter) involved in cognitive processing.(12,13) SSMs are based on the most dominant theory of how people make decisions:(14) Specifically, decisions (such as those in cognitive tests) are a result of processes (Figure 2: dotted lines) that have a starting point and that evolve by sequentially accumulating (noisy) evidence for response options (up to a criterion at which a choice is initiated  
(Figure 2). SSMs are well-established and account for behavior from a range of tests.(15,16) They utilize more information than conventional statistics (e.g., average reaction times) because parameters are derived from the simultaneous consideration of accuracy and the entire reaction time (RT) distributions for corrects and errors.(17,18)

Figure2_DDMpicture.png

Figure 2. The diffusion decision model

Picture 3.png

Joint-modeling approach

I use a hierarchical Bayesian framework(19,20) to establish a link between clinical, behavioral and physiological measures (Figure 3). For example: if one hypothesizes a direct link between physiological and behavioral data, the observed physiological patterns are then used to replace the model parameters. This creates a model with predictions about the behavioral data. There exist different possible linking functions;(19,20) each function assumes a different brain-behavior interaction and is tested as to how well it accounts for empirical data. I focus on gaussian process regression and single-trial logistic regression (machine learning algorithms) as possible linking functions.(20-23) 

Figure 3. Joint-modeling approach

References

  1. National Institute of Mental Health. The National Institute of Mental Health (NIMH) Research Domain Criteria (RDoC) initiative. https://www.nimh.nih.gov/research/research-funded-by-nimh/rdoc/index. shtml. Accessed March 22, 2021. 

  2. Wahlstedt, C., Thorell, L. B., & Bohlin, G. (2009). Heterogeneity in ADHD: Neuropsychological pathways, comorbidity and symptom domains. Journal of abnormal child psychology, 37(4), 551-564. 

  3. Ging-Jehli NR, Ratcliff R, Arnold LE. Improving neurocognitive testing using computational psychiatry-A systematic review for ADHD. Psychol Bull. 2021 Feb;147(2):169-231. doi: 10.1037/bul0000319. Epub 2020 Dec 28. PMID: 33370129. 

  4. Nigg, J. T., Willcutt, E. G., Doyle, A. E., & Sonuga-Barke, E. J. (2005). Causal heterogeneity in attention-deficit/hyperactivity disorder: do we need neuropsychologically impaired subtypes?. Biological psychiatry, 57(11), 1224-1230.  

  5. Nikolas, M. A., Marshall, P., & Hoelzle, J. B. (2019). The role of neurocognitive tests in the assessment of adult attention-deficit/hyperactivity disorder. Psychological assessment, 31(5), 685-698.  

  6. 6ennett D, Silverstein SM, Niv Y. The Two Cultures of Computational Psychiatry. JAMA Psychiatry. 2019 Jun 1;76(6):563-564. doi: 10.1001/jamapsychiatry.2019.0231. PMID: 31017638. 

  7. Huys QJM. Advancing Clinical Improvements for Patients Using the Theory-Driven and Data-Driven Branches of Computational Psychiatry. JAMA Psychiatry. 2018 Mar 1;75(3):225-226. doi: 10.1001/jamapsychiatry.2017.4246. PMID: 29344604. 

  8. Liu S, Dolan RJ, Heinz A. Translation of Computational Psychiatry in the Context of Addiction. JAMA Psychiatry. 2020 Jul 1. doi: 10.1001/jamapsychiatry.2020.1637. Epub ahead of print. PMID: 32609324. 

  9. Ratcliff R. A theory of memory retrieval. Psychological Review. 1978; 85:59–108. 

  10. Kofler, M. J., Irwin, L. N., Soto, E. F., Groves, N. B., Harmon, S. L., & Sarver, D. E. (2019). Executive functioning heterogeneity in pediatric ADHD. Journal of Abnormal Child Psychology, 47(2), 273–286. 

  11. Woods, S. P., Lovejoy, D. W., and Ball, J. D. (2002). Neuropsychological characteristics of adults with ADHD: A comprehensive review of initial studies. The Clinical Neuropsychologist, 16(1), 12-34. 

  12. Forstmann, B. U., Ratcliff, R., & Wagenmakers, E. J. (2016). Sequential sampling models in cognitive neuroscience: Advantages, applications, and extensions. Annual review of psychology, 67. 

  13. 13Turner, B. M., Palestro, J. J., Miletić, S., & Forstmann, B. U. (2019). Advances in techniques for imposing reciprocity in brain-behavior relations. Neuroscience & Biobehavioral Reviews, 102, 327-336. 

  14. Busemeyer, J. R., & Townsend, J. T. (1992). Fundamental derivations from decision field theory. Mathematical Social Sciences, 23, 255–282. 

  15. Logan, G. D., Van Zandt, T., Verbruggen, F., & Wagenmakers, E.-J. (2014). On the ability to inhibit thought and action: General and special theories of an act of control. Psychological Review,121(1),66-95.  

  16. Ratcliff, R., & Smith, P. L. (2004). A comparison of sequential sampling models for two-choice reaction time. Psychological review, 111(2), 333. 

  17. Van Zandt, T., & Ratcliff, R. (1995). Statistical mimicking of reaction time data: Single-process models, parameter variability, and mixtures. Psychonomic Bulletin & Review, 2(1), 20-54. 

  18. Ziegler, S., Pedersen, M. L., Mowinckel, A. M., & Biele, G. (2016). Modelling ADHD: a review of ADHD theories through their predictions for computational models of decision-making and reinforcement learning. Neuroscience & Biobehavioral Reviews, 71, 633-656.  

  19. Turner, B. M., Forstmann, B. U., Love, B. C., Palmeri, T. J., & Van Maanen, L. (2017). Approaches to analysis in model-based cognitive neuroscience. Journal of Mathematical Psychology, 76, 65-79. 

  20. Turner, B. M., Palestro, J. J., Miletić, S., & Forstmann, B. U. (2019). Advances in techniques for imposing reciprocity in brain-behavior relations. Neuroscience & Biobehavioral Reviews, 102, 327-336. 

  21. Turner, B. M., & Van Zandt, T. (2018). Approximating Bayesian inference through model simulation. Trends in Cognitive Sciences, 22(9), 826-840. 

  22. Turner, B. M., Van Maanen, L., & Forstmann, B. U. (2015). Informing cognitive abstractions through neuroimaging: The neural drift diffusion model. Psychological review, 122(2), 312. 

  23. Bahg, G., Evans, D. G., Galdo, M., & Turner, B. M. (2020). Gaussian process linking functions for mind, brain, and behavior. Proceedings of the National Academy of Sciences, 117(47), 29398-29406. 

bottom of page