If you have preregistered this abstract (e.g., on OSF) please share a link

You may upload a word copy of your abstract here (Please include your last name in the filename)

daniel.e.gustavson@vumc.org

Naomi Friedman, Mike Stallings, Reyna Gordon (Vanderbilt University)

https://ibg.colorado.edu/dokuwiki/doku.php?id=catslife:projects:start

First, I plan to examine the heritability of music in childhood (age 7-16 music interest and instruments played, and age 12 music aptitude). The primary analyses will focus on examining phenotypic and genetic correlations between childhood music interest/ability and later cognition (focus: executive function and language abilities), mental health (diagnostic interviews), and brain structure. Finally, I may include use data from parents (age 12) to examine gene-environment correlation and interactions.

CADD, LTS, CAP, CATSLife

Chandra Reynolds, Sally Wadsworth

https://ibg.colorado.edu/dokuwiki/doku.php?id=catslife:projects:start

The proposed research aims to test how individual differences including measured environmental experiences and genetic backgrounds shape the co-development of health salient traits, tobacco use, and cognition.

Shandell Pahlen, Chandra Reynolds

https://ibg.colorado.edu/dokuwiki/doku.php?id=catslife:projects:start

The proposed research aims to create a frailty index score for later adulthood, where frailty refers to multi-domain deficits in functioning and health

https://ibg.colorado.edu/dokuwiki/doku.php?id=catslife:projects:start

We will use a co-twin control study design to examine how cannabis use impacts various physical health outcomes.

CAP, LTS, CATSLIFE,CADD

Adriene Beltz, Sheri Berenbabum, Robin Corley

https://ibg.colorado.edu/dokuwiki/doku.php?id=catslife:projects:start

Compare prospective reports of status to a validated measure of retrospective pubertal timing using a large cohort of youth followed prospectively into adulthood

https://ibg.colorado.edu/dokuwiki/doku.php?id=lts_data:projects:in_progress:genetic_and_environmental_influences_on_stress_ef_internalizing_and_externalizing_behaviors:2018-09

I will be working with measures of stressful life events, executive function and externalizing and internalizing disorders. I aim to look at associations between dependent vs independent life events, executive function and externalizing / internalizing manifestations of psychopathology. By using twin data and a longitudinal design I hope to parse apart risk pathways for psychopathology via exposure to stressful life events and executive function deficits.

MA thesis defended, Manuscript drafting

Elizabeth Munoz, Chandra Reynolds, Paige Trubenstein, Robin Corley, Soo H. Rhee, Sally Wadsworth

https://ibg.colorado.edu/dokuwiki/doku.php?id=lts_data:projects:in_progress:parenting_stress_and_mental_health_in_midlife:2020-09

This study plans to examine the effects of parenting stress on parental mental health utilizing data from the ongoing CATSLife project. The study will include participants into the analysis sample if they reported having child(ren). We predict that higher levels of parenting stress will be associated with poorer reports of depressive and anxiety symptoms. We will also investigate whether neighborhood social resources, in the form of social capital, moderates this link.

MA Thesis; Chloe left project; Chandra, Ryan, others will reshape

Ryan Bruellman (newly added), Andrew Smolen, Luke M. Evans, Jarrod Ellingson, Naomi P. Friedman, Robin P. Corley, John C. DeFries, Sally J. Wadsworth, Chandra A. Reynolds

https://ibg.colorado.edu/dokuwiki/doku.php?id=catslife:projects:start

The proposed research aims to investigate serum Brain-derived neurotrophic factor or BDNF, and its association to cognitive performance, BMI, and physical activity, as well as its genetic architecture.

Poster presentation WPA

Fared (Fred) Dungore, Naomi Friedman, Robin Corley, Sally Wadsworth,
Dina Bach, Fady Lawrence, Seunghyeon (Mary) Lee, Thomas Lee

https://ibg.colorado.edu/dokuwiki/lib/exe/fetch.php?media=catslife:documents:reynoldslab_aofi_video_game_genre_impulsivity_inattention_draft4.docx

The proposed project will address the extent to which playing specific video game genres in CATSLife participants relates to impulsivity and the extent to which hours spent on videogames/computers/apps may mediate any associations. To address selection, we will account for age 16 impulsivity.

Chandra Reynolds, Robin Corley, Sally Wadsworth

https://ibg.colorado.edu/dokuwiki/doku.php?id=lts_data:projects:in_progress:activity_engagement_and_cognitive_trajectories:2015-10

This study will analyze activity engagement and cognitive abilities across multiple waves of data collection, starting with age 16 and CATSLife.

Elizabeth.Munoz@austin.utexas.edu

One Manscuript published, Other Analyses begun

Chandra Reynolds, Sally Wadsworth, Robin Corley, others

https://ibg.colorado.edu/dokuwiki/doku.php?id=lts_data:projects:in_progress:associations_between_environmental_characteristics_and_cognitive_function:2016-01

This study will examine the concurrent and enduring associations between exposure to stressful or negative environments and cognitive function.

https://ibg.colorado.edu/dokuwiki/doku.php?id=lts_data:projects:in_progress:sleep_cardiometabolic_pathways_to_cognitive_functioning:2018-07#

We will examine the relationship between self-reported sleep disruption, APOE status and AD PRS, cardiometabolic biomarkers including lipid profiles and BMI, and cognition using data from the CATSLife project.

Thersis completed; Updated analyses begun for publication

Chandra Reynolds, Shandell Pahlen, Robin Corley, Sergio Rey

https://ibg.colorado.edu/dokuwiki/doku.php?id=lts_data:projects:in_progress:activity_engagement_and_cognitive_trajectories:2015-10

Investigate rural-urban differences by considering proximal and distal rurality differences in leisure time activity engagement, social captial and cognitive performance

Andrew Smolen, Chris Link, Sally J. Wadsworth

https://ibg.colorado.edu/dokuwiki/doku.php?id=lts_data:projects:in_progress:activity_engagement_and_cognitive_trajectories:2015-10

The study will evaluate the neurofilament light chain (NfL) as a biomarker of cognitive functioning in early- to mid- adulthood

Analyses begun, Conference presentation

Elijah Knaap, Robin Corley, Kyle Gebelin, Elizabeth Munoz, Soo H. Rhee, Sally J. Wadsworth

https://ibg.colorado.edu/dokuwiki/doku.php?id=lts_data:projects:in_progress:activity_engagement_and_cognitive_trajectories:2015-10

Chandra Reynolds, Sergio Rey, Robin Corley, Sally Wadsworth

https://ibg.colorado.edu/dokuwiki/doku.php?id=lts_data:projects:in_progress:activity_engagement_and_cognitive_trajectories:2015-10

Park accessibility and physical health associations are evaluated, considering GIS park accessibility indices and Body Mass Index (BMI).

Fared (Fred) Dungore, Chloe Myers, Shandell Pahlen, Nick Bowman (UT Austin)

https://ibg.colorado.edu/dokuwiki/doku.php?id=lts_data:projects:in_progress:activity_engagement_and_cognitive_trajectories:2015-10

The coding and analysis will address the extent to which playing specific video games engages cognitive, social, emotional, and enjoyment/appreciation demands in participants, how these game features relate to cognitive performance.

Evaluate longitudinal geospatial models considering space-time context profiles: in-situ change vs residential change

capstone write-up, analyses

Shandell Pahlen, Robin Corley, Sally Wadsworth

https://ibg.colorado.edu/dokuwiki/lib/exe/fetch.php?media=catslife:documents:orozco_reynolds_aofi_capstone_abstract_ao-sjw-cr.docx

We will examine whether individuals approaching midlife with autoimmune conditions show reduced cognitive speed performance compared to those who do not report autoimmune conditions. Additional cognitive outcomes may be considered in the future.

Ryan Bruellman <rbrue001@ucr.edu>

Jarrod Ellingson, Shandell Pahlen, Robin Corley, Sally Wadsworth, Chandra Reynolds, Ilana Bennett, Luke Evans, Andy Smolen, Donald Evans

https://drive.google.com/file/d/18p4qLC_mzyHuEXz-nfiEghzXFnTpSs1g/view

We aim to evaluate if higher amounts of moderate to vigorous physical activity are supported for individuals who are otherwise sedentary throughout most of their regular week. This study will use participant data from CATSLife ranging from reported sitting time to reported physical activity and measured biomarkers. Sept 2025 update: Adding in biomarkers relevant to vascular inflammation (iCAM, vCAM, MCP1, etc.) that are available within CATSLife1 individuals through the NULISA assay.

10.1371/journal.pone.0308660

Ryan Bruellman <rbrue001@ucr.edu>

Eli Knaap, Sergio Rey, Robin Corley, Chandra Reynolds

https://ibg.colorado.edu/dokuwiki/lib/exe/fetch.php?media=catslife:documents:bruellman_abstract_2_-_food_access_food_quality_and_cost_cr-sjw.docx

We are interested to determine whether the presence of high-quality low-cost food sources including fruit stands and farmer’s markets correlate to higher fruit/veg and fiber scores in all individuals regardless of financial situation. We plan to also incorporate participant education level and neighborhood socioeconomic status as factors in food choice.

Analyses and writing begun

Chandra Reynolds, Sally J. Wadsworth, Naomi Friedman

https://ibg.colorado.edu/dokuwiki/lib/exe/fetch.php?media=catslife:documents:zheng_catslife_project_submission_cognitivereserve_lts.docx

Using a residual approach, we measured CR as the variance of cognitive performance unaccounted for by demographic variables and brain measures in LTS. To evaluate construct validity of CR, we we will evaluate measures such as impulsivity, conscientiousness, cognitive activity, etc, and outcomes such as a frailty index.

Chandra Reynolds, Naomi Friedman, Dan Gustavson, Robin Corley, Sally J. Wadsworth

https://ibg.colorado.edu/dokuwiki/lib/exe/fetch.php?media=catslife:documents:zheng_phenome-wide-cr0722.docx

Using three large-scale studies tapping on different life stages (ABCD: childhood, CATSLife: adulthood, VETSA: late-adulthood), we will investigate over 100 phenotypes on their potential role as CR proxies. We will use cross-validated machine learning models to conduct an atheoretical comparison across variables. We also plan to apply within-family design and Mendelian randomization to examine the direction of prediction between CR proxies and cognitive functioning.

hasm6947@colorado.edu

Naomi Friedman, Luke Evans, Colleen Reid

Utilizing address history data across the LTS and CATSLIFE, I plan to geocode participants' exposure to particulate air pollution and test for associations between exposure across the lifespan and neuropsychiatric outcomes in adulthood

CADD, GADD, LTS, CTS, CAP, CATSLife

Naomi Friedman, Chandra Reynolds

Study of psychosocial and cognitive risk factors for pain.

Elizabeth Muñoz, Robin Corley, Sally Wadsworth, and Chandra A. Reynolds

https://ibg.colorado.edu/dokuwiki/lib/exe/fetch.php?media=catslife:documents:choi_catslife_wiki_abstract_gsa_2022_revised_jc.docx

The current study will assess the independent and joint associations of objective (i.e., neighborhood socioeconomic deprivation) and subjective neighborhood characteristics (i.e., individuals’ perceived neighborhood stressors) on anxiety and depression among early midlife adults using data from the Colorado Adoption/Twin Study of Lifespan behavioral development and cognitive aging (CATSLife).

Deepika.Dokuru@colorado.edu

To date, the largest genetic risk factor identified in two diseases classified as tauopathies, PSP and CBD, is the H1 haplotype on chromosome 17. Although the exact process in which the tau haplotype may be involved in disease progression is unknown, the high frequency of the H1 haplotype among patients with PSP and CBD (and other neurodegenerative disorders) seems to implicate it as a risk factor of neurocognitive decline. To assess whether the risk arises from early differences/deficits in neurocognitive development, I want to assess differences in cognitive functioning at various ages that we have available cognitive data. So far, we have conducted these analyses in ABCD and found significant associations with fluid composite score (NIH Toolbox) and cortical area, and marginal significance with total composite score (NIH Toolbox). We hope to replicate these findings in the early CAP/LTS and CATSLife data set. Additionally, since APOE genotyping is available in this data set, we hope to incorporate that into our models to assess whether the effects of the haplotype persist over and above the effects of APOE.

DREW.WINTERS@cuanschutz.edu

Joseph Sakai, Christian Hopfer, Jarrod Ellingson, Robin Corley, Michael Stallings, John Hewitt, Tamara Wall

Conduct disorder (CD) is a child antisocial phenotype where 40% persist in these symptoms to develop antisocial personality disorder (ASPD) as an adult. Across the literature, CD associates with transdiagnostic symptoms of internalizing, externalizing, and impaired top-down cognitive processes (for review: Toro, 2020). Although it is well known that transdiagnostic symptoms are often comorbid, their association with the development of ASPD are less clear. One reason for this may be that studies in this topic area rarely account for the latent growth trajectory of transdiagnostic symptoms or the related top-down cognitive control processes. It is plausible that internalizing, internalizing, and top-down cognitive impairments are present in youth who develop adult ASPD and that changes in these symptoms could reveal important transdiagnostic associations with clinical relevance. The CADD, GADD, and LTS datasets are uniquely able to address the above inconsistencies and provide a significant contribution to this topic

Elisa.Stern@colorado.edu

Jarrod Ellingson, Jonathan Schaefer, Soo Rhee

CUD and depression and suicidality co-occur more frequently than would be expected by chance, and although direct causality remains unclear, cannabis has been proposed as a plausible causal mechanism for mood maladjustment.The present study used a sibling-comparison design in a sample of high-risk adolescents, in which one sibling was considered a clinical proband, having been identified due to delinquent behaviors. First, analyses examined the association between cannabis use and depression and suicidality. Second, multilevel models partitioned phenotypic associations into between family (i.e. familial factors shared by siblings) and within-family (i.e. differing levels of cannabis use between siblings) effects. A final multilevel model accounted for alcohol use.

Harry.Smolker@colorado.edu

Naomi Friedman, Marie Banich

While studies have frequently shown strong associations between increased severity of depression and impairments in executive functioning (EF), the degree to which this relationship is grounded in common neural mechanisms remains widely unexplored. This study aims to investigate the neuroanatomical correlates of executive function (EF) and depressive symptomology, focusing on the degree to which EF impairments and depressive symptomology are related to common or distinct neural substrates. Using a range of neuroanatomical measures, including cortical volume, thickness, surface-area, and gyrification, as well as fractional anisotropy of white matter tracts, we will 1) identify the neuroanatomical correlates of distinct constructs of EF, 2) identify the neuroanatomical correlates of a range of symptoms of depression, and 3) search for brain regions with neuroanatomical properties that mediate EF/depression relationships. Additionally, analyses taking into account twin status will investigate the degree to which brain-behavior relationships are influenced by genetic and environmental factors.

MEGAN.ROSS@cuanschutz.edu

Several studies have evaluated the association between cannabis use and BMI, however, results have been mixed. This study will analyze the longitudinal association between cannabis use and BMI among adolescents.

katie.paulich@colorado.edu

John Hewitt, Michael Stallings, Deepika Dokuru, Sam Fries

There has been little or no research conducted on relationships between internalizing problems (e.g., depression, anxiety, self-esteem) and risky sexual behavior in twins. The present research aims to close this gap. Externalizing problems will also be examined in regard to risky sexual behavior. Three publications should come out of this research: internalizing and risky sexual behavior, externalizing and risky sexual behavior, and both internalizing/externalizing and risky sexual behavior.

katie.paulich@colorado.edu

I aim to examine potential relationships between PTSD and risky sexual behavior (RSB). I may also examine childhood abuse links to RSB as well, as PTSD may be a result of such abuse. I intend to use a twin sample, but have not finalized details.

Causal Mechanisms Underlying the Association Between Socioeconomic Status and Antisocial Behavior: A Longitudinal Examination of The Colorado Adoption Study

The aims of this study are to (1) replicate previous findings and test whether environmental mediation, gene–environmental correlation or a combination of the two processes influence the association, (2) examine the role of gene–environment interaction in the association, (3) examine the influence of age in the association, (4) test for sex differences in the association, and (5) test for differences in the magnitude of associations due to environmental mediation and gene–environment correlation for aggressive and non-aggressive antisocial behavior. Participants are adoptive, biological, and control families from the Colorado Adoption Study.

lauren.michelle.bertin@emory.edu

Naomi Friedman, Rohan Palmer

Substance use disorders often co-occur with other psychopathologies. In particular, individuals diagnosed with conduct disorder (CD) during adolescence are nearly five times more likely to develop alcohol use disorder (AUD [i.e., DSM symptom count]). Similarly, the literature indicates that individuals with attention deficit hyperactivity disorder (ADHD) are also at higher risk for AUD. Researchers have suggested that these childhood/adolescent externalizing behaviors should be considered as premorbid indicators of alcohol misuse. The link between externalizing behaviors and AUD has previously been examined using CADD data (Palmer, et al. 2013), but not in the context of executive dysfunction, and not beyond young adulthood. Moreover, the literature on the prospective effects of adolescent externalizing on alcohol use is limited and at times mixed. Recent studies suggests that both AUD and ADHD/CD share common underlying factors: reduced self-restraint and deficits in executive functioning (EF), suggesting that these intermediary phenotypes may offer greater sensitivity to detecting risk for alcohol misuse beyond the early years of experimentation and initial use. We propose to examine whether distinct developmental trajectories in toddler self-restraint (i.e., class membership) predict future alcohol initiation and patterns of use (i.e., symptoms and frequency). In particular, we will compare different types of parallel process growth models that utilizes class membership as a predictor of all three waves of alcohol use (i.e., frequency/quantity) and disorders. While the CADD data has been used to examine the relationship between toddler self-restraint and executive functioning (Friedman, Miyake, Robinson, & Hewitt, 2011), as well as toddler self-restraint and later externalizing problems (Rhee, et al. 2018), the relationship between toddler self-restraint and substance (in this case alcohol) use has not yet been investigated. We will examine the relationship between EF and alcohol use over the three waves. While this has not yet been conducted, findings from Gustavson et al. (2017) on the relationship between polysubstance use and common EF suggest that there may be a prospective association present in CADD data. Running the proposed models will clarify how alcohol use and EF affects are prospectively inter-related. We also propose to expand upon the two previously described models. These will be combined so that toddler self-restraint can be examined as an early indicator of the relationship between alcohol use and EF. As mentioned above, our analyses will account for known behavioral correlates of substance use and EF (i.e., ADHD, CD and NS) by including them as covariates in the models. Lastly, since the relationship between self-restraint and EF appears to biological in nature (i.e., driven by common genetic factors), we will then generate polygenic risk scores (PGRS; i.e, for alcohol, ADHD, CD) from an external dataset and use them to predict the relationship between alcohol use and executive functioning, and compare whether self-restraint membership or these polygenic risk scores explain the patterns observed better. In other words, detecting shared genetic effects between traits of interest in our study by using PGRSs, will allow us to understand whether directly measuring self-restraint is necessary to predict future alcohol misuse.

luke.m.evans@colorado.edu

Following Genomic SEM analyses of internalizing and alcohol use phenotypes, we plan on using the CADD/GADD samples as a target sample for prediction analyses. Using the multivariate Genomic SEM GWAS SNP effects for each of the latent factors to build prediction weights and predict alcohol use and internalizing phenotypes in the CADD/GADD sample.

Maia.Frieser@colorado.edu

Following up on a previous project, I plan to do a discordant twin design to examine any potential causal relationships between cannabis use in adolescence and educational attainment, using measures of frequency of cannabis use and educational attainment. I will use variables such as highest level of education completed and twin data to examine this relationship.

stephanie.zellers@helsinki.fi

I will investigate whether environmental permissiveness (based on year of assessment and state-level marijuana trends) is related to the relationship between behavioral disinhibition and marijuana use. Behavioral disinhibition is modeled as a latent factor for which the manifest variables are attention deficit hyperactivity disorder, conduct disorder, and alcohol dependence. Data from CADD will be harmonized with data from the Minnesota Twin/Family Study.

Tor.D.Wager@dartmouth.edu

Marta Ceko, Naomi Friedman, Matt Keller, John Hewitt

This project investigates brain mechanisms of placebo analgesia, combining neuroimaging, behavioral, and genetic approaches. We will use fMRI to characterize the neural bases of placebo effects in 600 twins recruited from the Colorado Twin Sample and predict individual differences in placebo effects across two forms of pain. In Aim 1, we will develop models that predict the magnitude of individuals’ placebo effects in pain and pain neurophysiology based on a) fMRI activity, b) brain structure, and a combination of personality, behavioral, and cognitive measures that can be deployed clinically. In Aim 2, we conduct analyses of heritability of placebo effects and their neural predictors, and genetic correlations that can identify brain features whose relationships with placebo effects are genetic in origin. In Aim 3, we leverage the >50,000 person ENIGMA consortium to identify genome-wide associations with placebo-linked brain features and develop polygenic risk scores for placebo effects.

Yongkang.kim@colorado.edu

Matt Keller, Jared Balbona, Deepika Dokuru

We are planning to perform genome-wide association studies (GWAS) based on siblings and MZtwins dataset as the part of family GWAS consortium.

Elisa Stern, Soo Rhee, Chandra Reynolds, Naomi Friedman

The construct of internalizing captures shared variance underlying risk for mood and anxiety disorders (sometimes including post-traumatic stress disorder [PTSD]). Internalizing is typically modelled with latent factors capturing diagnoses (e.g., based on DSM criteria for major depressive disorder [MDD] or generalized anxiety disorder [GAD]) and/or self-reported questionnaires tapping depression and anxiety symptoms and neuroticism personality traits. The diathesis-stress model suggests that the underlying genetic liability for psychopathology is continuously distributed, suggesting that these sets of measures should capture largely the same set of genetic risk factors (and potentially similar environmental risk factors). Here, we propose two sets of analyses that will evaluate this possibility. First, using genomic structure equation modeling, we will construct internalizing factors based on genome-wide association studies (GWAS) of internalizing traits based on either diagnoses (e.g., MDD, GAD, PTSD) or self-reported symptoms (neuroticism, loneliness, reverse-scored subjective well-being) and test the hypothesis that the genetic correlation between these factors should be close to 1.0. Second, we will construct similar factors based on measures available in CATSLife (GAD and MDD symptom counts vs. questionnaire measures of neuroticism, worry, rumination, and anxiety/depression symptoms). Leveraging twin and adoptive data from CATSLife, we will then examine the genetic and environmental correlations among internalizing factors using a full ACDE model, again testing the hypothesis that the correlations should be near 1.0 (especially for genetic factors). This work therefore provides an empirical test of the assumptions of the diathesis-stress model, and will improve our understanding of the genetic and environmental influences captured by different measures of internalizing.

The construct of internalizing captures shared variance underlying risk for mood and anxiety disorders (sometimes including post-traumatic stress disorder [PTSD]). Internalizing is typically modelled with latent factors capturing diagnoses (e.g., based on DSM criteria for major depressive disorder [MDD] or generalized anxiety disorder [GAD]) and/or self-reported questionnaires tapping depression and anxiety symptoms and neuroticism personality traits. The diathesis-stress model suggests that the underlying genetic liability for psychopathology is continuously distributed, suggesting that these sets of measures should capture largely the same set of genetic risk factors (and potentially similar environmental risk factors). Here, we propose two sets of analyses that will evaluate this possibility. First, using genomic structure equation modeling, we will construct internalizing factors based on genome-wide association studies (GWAS) of internalizing traits based on either diagnoses (e.g., MDD, GAD, PTSD) or self-reported symptoms (neuroticism, loneliness, reverse-scored subjective well-being) and test the hypothesis that the genetic correlation between these factors should be close to 1.0. Second, we will construct similar factors based on measures available in CATSLife (GAD and MDD symptom counts vs. questionnaire measures of neuroticism, worry, rumination, and anxiety/depression symptoms). Leveraging twin and adoptive data from CATSLife, we will then examine the genetic and environmental correlations among internalizing factors using a full ACDE model, again testing the hypothesis that the correlations should be near 1.0 (especially for genetic factors). This work therefore provides an empirical test of the assumptions of the diathesis-stress model, and will improve our understanding of the genetic and environmental influences captured by different measures of internalizing.

https://drive.google.com/open?id=1cnItQung_j_caX9vP_eupirO2mmartl8

Elizabeth Muñoz, Anqing Zheng, Sally J. Wadsworth, Martin J. Sliwinski, Chandra A. Reynolds

I am uploading an abstract to the projects in progress WIKI which aims to examine micro sleep and cognition associations across 14 days using the CATSLife1 smartphone data. This abstract outlines my proposal for this examination and will be the focus of the second chapter of my dissertation. This abstract has been circulated to the co-authors.

https://drive.google.com/open?id=1e5175qRp4-P93Ww2q6fFfo1f-oYCNu61

Chandra A. Reynolds, Michael C. Stallings, Daniel Gustavson, Robin P. Corley, Sally J. Wadsworth

Individual differences in personality and cognition play a fundamental role in how people interact with the world, organize their introspective perspectives, and form a central part of their identities. These characteristics are also interrelated (e.g., Wettstein et al., 2017; Sutin et al., 2022), and may partially originate from genetic pleiotropy through complex interactions between genetics and environment (Hindley et al., 2022). However, these measured genetic variants only capture a miniature proportion of the variants, and the trajectory of shared genetic effects on the relationship between personality and cognition across development remains unclear. Additionally, it is important to consider the interindividual heterogeneity in both personality and cognition, as different patterns of genetic and environmental contribution may emerge across different personality-cognition dynamics. By utilizing a genetically informed sample (i.e., twins, biological and adoptive siblings; CATSLife), we will explore three research questions: 1) what is the magnitude of association and variation between covariance of Big Five personality items and specific cognition domains? 2) To what extent do genetic and environmental effects contribute to the covariance of personality and cognition? 3) What is the trajectory of the genetic and environmental effects across early to middle adulthood? Using zero-order network analyses, we were able to organize the association across Big Five personality items and cognition domains. For cognition domains, we included measures of episodic memory, processing speed, spatial ability, and vocabulary. Synthesizing these results with the existing body of research on the dynamics of personality-cognition will help us expand and refine our understanding of the complex nature of psychological traits.

https://drive.google.com/open?id=1TmUUju2gYaQnYLmBT5k7MpCFG2TFGl1I

Tina T. Vo, Elizabeth Muñoz, Sally J. Wadsworth, Martin J. Sliwinski, Chandra A. Reynolds

Calibration accuracy, the agreement between individuals' perceived performance and actual performance, plays a critical role in self-regulated learning, knowledge monitoring, and goal maintenance. However, it is unclear how calibration accuracy operates at the micro-level timescales, especially among individuals approaching midlife. We investigated the convergence between perceived cognitive performance and the trajectory of momentary assessments collected repeatedly over 14 days. We analyzed the data of 440 participants (Mage = 35.78, %Female = 57.7%) from the Colorado Adoption/Twin Study of Lifespan Behavioral Development and Cognitive Aging (CATSLife1), who completed spatial working memory tasks administered via smartphones up to three times a day. On average, participants' working memory improved across the 14-day period (β=-.014, SE=.001, p<.001), with lower scores indicating more accurate placement and better performance. Moreover, we observed less day-to-day variability in working memory performance (β=-.010, SE=.001, p<.001). The trajectory of working memory performance was consistent with participants' evaluations of cognitive performance after completing the study: those perceiving improvement in their performance tended to perform better overtime (r = .25, SE = .046) and showed less variability in momentary assessments (r = -.28, SE = .046). Participants' self-reported performance was also positively correlated with their working memory intercept at the start of the study (r = .34, SE=.045). Exploratory analyses then examined the associations between momentary calibration accuracy and other personality characteristics (e.g., openness). Our findings suggest that understanding the dynamic interplay between calibration accuracy and cognitive performance could shed light on individual differences during periods of cognitive maintenance versus decline.

https://drive.google.com/open?id=11GCs3vwZizlElQGUX1b3CsFMhHj3hDbA

zoe.panchal@cuanschutz.edu

Joseph Sakai, Kenneth Wright, Christian Hopfer, Megan Ross (senior author), John Hewitt, Matt McGue, William Iacono, Scott Vrieze, Robin Corley

The primary aim of this project is to examine the association between recreational cannabis legalization and use of cannabis as a sleep aid. In addition, we will examine the association between recreational cannabis legalization and use of sleep medication and alcohol as sleep aids.

daniel.gustavson@colorado.edu

Chandra Reynolds, Naomi Friedman, Hermine Maes (VCU), Reyna Gordon (VUMC), Sally Wadsworth, Mike Stallings

Studies on the genetics of musical ability and music engagement are few and far between, but our recent work suggests self-reported music interest and skills are highly heritability by age 12 in CATSLife. We plan to further investigate the longitudinal trajectories of multiple music engagement measures in CATSLife, including the extent to which genetic and environmental influences on music engagement change between childhood (age 7) and adolescence (age 17) across 5 assessments. First, we will evaluate competing phenotypic models (e.g., common factor, autoregressive, linear and variable growth models) of music measures assessed in LTS and CAP samples (mean ages 7, 9, 10, 12, and 16). Second, using the best-fitting models, we will fit twin/family models to decompose variance into ACDE influences (e.g., individual differences in intercept and slope latent factors). Third, we will examine how these measures predict music engagement phenotypes in CATSLife wave 1 (i.e., weekly hours of music listening or music playing). In addition, we will use an extended family design to examine genetic and environmental influences on music engagement at the age 12 assessment (i.e., including mother-reported and father-reported music interest and talent, as well as music test data for children, mothers, and fathers available in CAP only). This final set of analyses may be broken off into a separate journal article depending on the complexity of the analyses and feedback from authors. Dan Gustavson will primarily collaborate with Dr. Hermine Maes (VCU) on these analyses, but will request feedback from all IBG authors and provide regular updates to Chandra Reynolds and Naomi Friedman. Dr. Maes may assist with some analyses, but Dr. Gustavson will conduct most analyses (potentially with graduate student assistance). This work will improve our understanding of genetic and environmental influences on early music engagement and the extent to which music engagement later in adolescence and adulthood is influenced by similar/unique sets of genetic/environmental influences. Music engagement is related to important cognitive abilities (language, executive function) and may be related to positive mental health outcomes, so this longitudinal work sets the stage for future projects examining how these trajectories map onto those for cognition and health.

https://drive.google.com/open?id=1j-AamlXLBKELX1ehX_99cC9Xe1Vw62ZB

michael.stallings@colorado.edu

In this experiment, we plan to use the CADD and GADD high risk probands to investigate the extent to which increased access to cannabis products containing high levels of THC concentration are associated with more severe consequences of cannabis use. The CADD and GADD samples provide us a natural experiment because all of the CADD probands were tested 3 times prior to January 2014 (the date cannabis was legalized in Colorado) while the GADD probands were only tested twice prior to January 2014 and once afterwards. Thus, the CADD sample provides us with a normative cannabis use impairment trajectory in a high-risk sample over three developmental periods including adolescence, emerging adulthood, and adulthood. We will compare this with the GADD sample which has the environmental event of cannabis being legalized in Colorado occurring between wave two (emerging adulthood) and wave three (adulthood) assessments. Surprisingly, current research has shown that the increase of access to cannabis products containing high levels of THC concentration has not led to an increase in Cannabis Use Dependence (CUD). Therefore, our hypothesis is that although the prevalence of CUD has remained constant, those who meet the criteria for CUD are suffering greater consequences.

https://drive.google.com/open?id=1fuG_Dh8aJ7GShXYiMeeTtpE4R6nXa-4u

Heylee.Hewawasam@colorado.edu

This project uses the Living Arrangements data collected in waves 2 and 3 of the GADD probands. We hypothesize that V-shaped trajectories (high at wave 1, low at wave 2, high at wave 3) will show greater incarceration, treatment, or other restrictive conditions that interrupt the patterns of regular users. The inverse trajectory (low at wave 1, high at wave 2, low at wave 3) will show normative aging out - due to factors such as employment, marriage, and family - but may also show higher levels of restrictive conditions. Treatment will be associated with longer lasting periods of decreased use than incarceration.

https://drive.google.com/open?id=1jlP8R-723GF8nY_n87KkQuj-Vgu5yOp0

Dan Gustavson, Soo Rhee, Emma Skoler

Negative emotionality is an important trait to study because high levels of negative emotionality are associated with negative mental outcomes, such as mood and eating disorders (Malouff, Thorsteinsson, & Schutte, 2005), emotion dysregulation, and psychological inflexibility (Paulus et al., 2016). Moreover, understanding how negative emotionality is passed from parent to child may inform the development of interventions designed to increase emotional stability in children.

Although there is some evidence for heritability of negative emotionality, how negative emotionality is passed from parent to children through parenting traits and environmental influences is unclear. To test alternative hypotheses regarding the transmission of negative emotionality from parents to children, we will compare the magnitude of the association between negative emotionality in parents and their children in biological and adoptive families. Specifically, we will evaluate whether these associations are consistent with environmental mediation (i.e., parents’ negative emotionality having a direct influence on their children’s negative emotionality), or passive gene–environment correlation (i.e., parents pass along both genes predisposing them to higher levels of negative emotionality and negative environmental influences correlated with those genes). Additionally, we will address whether specific parenting traits (i.e., negativity and warmth) mediate the parent–child similarity in negative emotionality. Lastly, we will examine evidence for evocative gene–environment correlation by examining whether children with higher levels of negative emotionality elicit more negative parenting and whether children who are more genetically similar receive more similar parenting. We will examine data from the Colorado Adoption Project (CAP; Plomin & DeFries, 1983) to address these questions.

https://osf.io/vgb3p/?view_only=b8c9b5d60b4d4dfe9e61b2cd5f364390

J. Megan Ross, Christian Hopfer, Mike Stallings, Tammy Wall

This study will assess the relationship between cannabis use and cognition in adolescents from GADD with ADHD by comparing three groups (ADHD only, cannabis use only, ADHD + cannabis use) on measures of cognitive function (inattention, executive function, and memory). We hypothesize that cognitive function will be worse in the ADHD + cannabis use group compared to either alone.

https://drive.google.com/open?id=1rnvnXK9lHt0qBS6xjX1B7pW8wTUmo38M

Ross, Hopfer, Vrieze, Hewitt, McGue, Ellingson, Sakai

Compare past month users of cannabis as a sleep aid to those who did not use cannabis as a sleep aid in the past month on relevant demographic, substance use and mental and physical health characteristics, controlling for sleep quality and frequency of past 6 month cannabis use. Specific outcomes of interest include: sex, race, age, income, education, problematic cannabis use, problematic alcohol use, use of alcohol as a sleep aid, use of medication as a sleep aid, mental health concerns (Personality Inventory for DSM-5 scores), self-rated physical health, and cognitive function (ICAR).

https://drive.google.com/open?id=1MEZ8Ao1_HKpGuVAXh_0qxUE-_44u5vHl

Assess the link between depression/suicidality and cannabis use in adolescents with ADHD,
by comparing 3 groups on depression and suicidality endorsement: ADHD + cannabis use vs ADHD only vs cannabis use only.

https://drive.google.com/open?id=170Rz0090_IcEWwMUrsdBFreFECUCIw0a

Robin Corley, Christian Hopfer, John K. Hewitt, Jacob Anderson, Scott Vrieze, Matt McGue, James J. Lee

Cognitive skills and non-cognitive skills (such as saving disposition) correlate with socioeconomic outcomes. However, given the complexity of social interactions, it is uncertain whether these associations represent causal effects. In the proposed study, we aim to identify causal effects on financial distress. We choose to examine some of the most potent predictors of socioeconomic outcomes: education, income, cognitive ability, and saving disposition. We model the effect of each one on financial distress, in a genetically-informed sample of 3,920 American twins. We use the family structure of our sample to control for familial (genetic or environmental) confounding. To ensure that the identified effects are causal, we employ polygenic scores as instrumental variables, in a within-family Mendelian randomization design

https://drive.google.com/open?id=1nTpCsW2DNJD3c5M81Clt2u3vUHECyZVr

Hugo Cogo Moreira, Ali Nouri, Sabine Pompeia, Daniel Gustavson, Naomi Friedman

Executive Functions (EF) are a set of cognitive abilities that control behavior to reach non-automatized goals people have in mind. One of the most widely accepted theoretical account of EFs is the EF unity and diversity framework: a pattern of intercorrelation (unity) among three domains of EF (inhibition, shifting, updating) that is not near perfect, so indicating separability (diversity) of these three types of EF. These domains are measured as latent factors using Structural Equation Modeling (SEM). Despite being highly popular, the configural structure of this framework has been under debate due to various contentions among researchers. Among other issues, the debate includes difficulties in dissociating executive abilities from lower-level cognitive skills (LLCS: e.g. sensory, naming, psychomotor) which are also recruited to carry out EF tasks. Most studies address this by using executive cost measures (i.e. performance on trials of tasks that recruit EF plus LLCS controlled for performance on similar trials that however only involve LLCS, but not EF). This, however, is only consistently done in the literature for inhibition and shifting, but not for updating tasks. Hence, it is unclear to what extent LLCS impact EF unity/diversity model configuration. This issue was explored in 407 9-15-year-olds (170 girls) from a developing country (Iran). The EF test battery included two tasks per EF domain to form latent factors. We tested three models using SEM in which indicators were raw scores (accuracy divided by RT to control for speed-accuracy tradeoffs): 1) an oft-found three-intercorrelated latent factor configuration using as indicators scores of trials with executive requirements only (not EF costs); 2) the same model with the addition of a latent factor of LLCS (performance on trials of the inhibition and shifting tasks with no executive requirements) regressed on the three EF latent factors. Model 1 showed high intercorrelation (0.82 to 0.96) between the three EF domains, replicating prior unity/diversity findings in studies that controlled for LLCS only on inhibition and shifting tasks. In model 2, LLCS highly predicted (βs > 0.90) all three EF latent factors which, however, were no longer significantly intercorrelated. We conclude that the ‘unity’ account of EFs is a potentially artificial pattern of intercorrelations that no longer holds once LLCS is accounted for in all three EF latent domains.

https://drive.google.com/open?id=1NQbeYWRUz_onokU0sss-mXdTW9LssYMX

harry.smolker@colorado.edu

Naomi Friedman, Luke Evans, Colleen Reid

A growing body of evidence suggests exposure to fine particulate air pollution (PM2.5) is associated with increased risk for mental illness, potentially years into the future. However, such studies have been limited in their longitudinal scope, generally testing for associations between exposure and psychopathology outcomes less than a decade following exposure. As a result, critical questions remain as to the long-term impacts of PM2.5 exposure, including the degree to which exposure during key developmental periods may affect psychopathology in adulthood, up to three decades following initial exposure. In the current project, we utilize data from the Longitudinal Twin Study (LTS) and the Colorado Adoption/Twin Study of Lifespan Behavioral Development and Cognitive Aging (CATSLife) to test for associations of estimated PM2.5 exposure at multiple developmental timepoints (ages 0, 6, 12, 17) with various psychiatric outcomes across adolescence and into adulthood (ages 12, 17, 23, 28 and older). PM2.5 exposure will be estimated by integrating participants’ address history data with a pre-existing dataset of estimated historical annual average levels of ambient PM2.5 across the United States, at a resolution of 1km2 (Meng et al., 2019, Environmental Science and Technology), while psychopathology will be measured both dimensionally and categorically, utilizing the Child Behavior Checklist and Kiddie Schedule for Affective Disorders and Schizophrenia. Notably, this unique dataset will allow us to investigate several critical issues related to the longitudinal impacts of PM2.5 exposure on psychopathology, including the degree to which there are specific sensitive periods or trajectories of exposure associated with particularly heightened long-term effects, whether exposure at different ages effects distinct aspects of psychopathology, and whether exposure may alter the trajectory of psychopathology over time. In addition, we will evaluate the degree to which associations between exposure and psychopathology differ between the sexes and are moderated by an individual’s exposure to psycho-social stressors.

not currently pre-registered

harry.smolker@colorado.edu

Naomi Friedman, Luke Evans, Colleen Reid

In addition to well documented effects of fine particulate air pollution (PM2.5) on cardiovascular and respiratory systems, recent evidence suggests PM2.5 may directly impact the brain, with potentially long-lasting impacts on mental health and cognition. However, it remains unclear if developmental PM2.5 exposure is associated with alterations in brain structure and organization in adulthood, and whether such effects may drive long-term associations between exposure and psychopathology. In the current project, we utilize data from the Longitudinal Twin Study (LTS) and the Colorado Adoption/Twin Study of Lifespan Behavioral Development and Cognitive Aging (CATSLife) to test for associations of estimated PM2.5 exposure at multiple developmental timepoints (ages 0, 6, 12, 17) with multiple properties of the adult brain. PM2.5 exposure will be estimated by integrating participants’ address history data with a pre-existing dataset of estimated historical annual average levels of ambient PM2.5 across the United States, at a resolution of 1km2 (Meng et al., 2019, Environmental Science and Technology), while neural measures will include gray matter morphometry, diffusion tensor imaging, and functional connectivity. Of particular interest is determining whether exposure to PM2.5 at distinct developmental timepoints have differential long-term effects on the adult brain and whether such effects may differ according to sex.

not currently preregistered

harry.smolker@colorado.edu

Naomi Friedman, Luke Evans, Colleen Reid

To accurately characterize the mechanisms by which air pollution affects mental illness, it is important to identify specific genetic pathways that may be involved. Whereas genetics appear to influence the effects of PM2.5 exposure on cardiovascular and respiratory health, it remains unclear if genes also influence the effects of PM2.5 exposure on neuropsychiatric outcomes. The current proposal utilizes pathway-specific polygenic risk scores (PRS) to address this question, testing for moderating effects of genetic risk for psychiatric traits of interest in specific genetic pathways on associations between exposure and neuropsychiatric traits. We specifically focus on gene pathways involved in neuroinflammation, including those supporting peripheral inflammation and cytokine production, microglial activation, and antioxidant activity. Additionally, we plan to use Genomic SEM to determine if there are any psychiatric-specific genetic risk factors, over and above the genetic risk that is shared between psychiatric conditions and cardiovascular and respiratory conditions. This information may help identify individuals who are at heightened genetic risk for adverse effects of PM2.5 exposure, while highlighting specific biological pathways that may serve as targets for exposure remediations.

not currently preregistered

Melissa A. Munn-Chernoff, Kari E. North, Cynthia M. Bulik, Heather M. Highland, Annie Green Howard, Soo Hyun Rhee, Michael Stallings, Robin Corley, John Hewitt, Daniel Gustavson

Obesity [defined as body mass index (BMI) ≥ 30kg/m2] is a major global public health concern, causing extensive burden on the healthcare systems.1 Previous studies have reported phenotypic associations between obesity with addictive behaviors, such as binge eating, alcohol use (i.e., number of drinks per week), and cigarette smoking (i.e., number of cigarettes smoked per day). For example, a strong positive association has been established between binge eating disorder and obesity.2 Results regarding phenotypic associations between cigarette smoking and alcohol use with BMI have been mixed. Positive phenotypic associations exist between BMI and number of cigarettes smoked,3, 4 yet negative associations have also been found between number of cigarettes smoked with BMI.5 In terms of alcohol use, some studies have found little to no associations with BMI in men and women, yet other studies have found positive associations between alcohol use with BMI in men and women.6 Further understanding the associations between addictive behaviors with obesity and exploring the mechanisms underlying these association is necessary to better understand the etiology of obesity and obtain a better characterization of individuals who are more likely have co-occurring health risks.

Genetics is a promising mechanism to explicate the addictive-like process in the co-occurrence of binge eating, alcohol use, smoking, and obesity given its potential to clarify molecular functions and identify novel biologic pathways. Twin studies and genome-wide association studies (GWAS) have demonstrated pleiotropy between binge eating, smoking, alcohol use, and BMI. For example, a twin study has demonstrated a moderate genetic correlation (r=0.34) between binge eating with BMI.7 A GWAS has reported a positive genetic correlation of 0.58 between alcohol use disorder and smoking, as well as a negative genetic correlation of -0.35 between alcohol use and BMI.8 However, no study to date has investigated the shared genetics among these traits in the same sample or longitudinally.

We propose to conduct a longitudinal twin analysis to further examine shared genetics between binge eating, alcohol dependence, smoking, and BMI, using data from participants in the Community Twin Study and Longitudinal Twin Study from the Colorado Center for Antisocial Drug Dependence (CADD). The CADD provides comprehensive measures for addictive behaviors among adolescents at multiple time points, allowing us to examine shared genetics among addictive behaviors and obesity longitudinally. We would like to include the following measures: Self-Report Questionnaire/Health Attitudes at Waves 1-3, which assessed height, weight, and other demographic information;9 Monitoring the Future Survey at Waves 1-3, which assessed cigarette and alcohol consumption;10 Eating Disorder Examination Questionnaire at Wave 3, which assessed eating disorder pathology;11 and the 7-item screening for eating pathology at Wave 2.12

References:
1. Apovian, C.M., Obesity: definition, comorbidities, causes, and burden. Am J Manag Care, 2016. 22(7 Suppl): p. s176-85.
2. Aguera, Z., et al., A review of binge eating disorder and obesity. Neuropsychiatr, 2021. 35(2): p. 57-67.
3. Wills, A.G. and C. Hopfer, Phenotypic and genetic relationship between BMI and cigarette smoking in a sample of UK adults. Addict Behav, 2019. 89: p. 98-103.
4. Bertoni, N., et al., Assessing the relationship between smoking and abdominal obesity in a National Survey of Adolescents in Brazil. Prev Med, 2018. 111: p. 1-5.
5. Jacobs, M., Adolescent smoking: The relationship between cigarette consumption and BMI. Addict Behav Rep, 2019. 9: p. 100153.
6. Traversy, G. and J.P. Chaput, Alcohol Consumption and Obesity: An Update. Curr Obes Rep, 2015. 4(1): p. 122-30.
7. Bulik, C.M., P.F. Sullivan, and K.S. Kendler, Genetic and environmental contributions to obesity and binge eating. Int J Eat Disord, 2003. 33(3): p. 293-8.
8. Kranzler, H.R., et al., Genome-wide association study of alcohol consumption and use disorder in 274,424 individuals from multiple populations. Nat Commun, 2019. 10(1): p. 1499.
9. Jessor, R. and S.L. Jessor, Problem behavior and psychosocial development: A longitudinal study of youth. (No Title), 1977.
10. Bachman, J.G., L.D. Johnston, and P.M. O'Malley, Monitoring the future. Ann Arbor: University of Michigan, Institute for Social Research, 1987.
11. Fairburn, C.G. and S.J. Beglin, Eating disorder examination questionnaire. Cognitive behavior therapy and eating disorders, 2008. 309: p. 313.
12. Munn, M.A., et al., Bivariate analysis of disordered eating characteristics in adolescence and young adulthood. Int J Eat Disord, 2010. 43(8): p. 751-61.

https://drive.google.com/open?id=10MGXG908ycSOxJ9LH8FU0pZJFXaKcRTH

luke.m.evans@colorado.edu

CADD, GADD, LTS, CTS, CAP, CATSLife, COTWINS

Marisa Lange, Emmanuel Sapin

Individual measures of the number of crossing-over events per meiosis are estimable from parent-offspring genetic data. Emmanuel has a pipeline to count the number of crossing-over events given phased haplotype data from parents and offspring. We are running a GWAS of the count of crossing-over events per parent to identify genetic variants, genes, and gene sets associated with individual-level variation in recombination rate, with a goal of evaluating genetic correlation and prediction of other health outcomes from polygenic scores. We note that several such studies have been done, mostly relying on Icelandic decode data, but none recently. This is an opportunity to use local data alongside several other public datasets to increase sample sizes and improve power to detect associations.

https://drive.google.com/open?id=1jXTi4uh-jPh7GIVpNfzeZZn2cJVWljOu

moka6421@colorado.edu

https://drive.google.com/open?id=1fxxMY2E-F28FHBhmCcy6ZQBWMs3EFPdb

chloe.mcghee@colorado.edu

By leveraging the LTS sample, we plan to investigate whether attention problems reported in childhood and early adolescence predict deficits in common executive function and impulsivity in early adulthood (age 23 and/or age 28). Within the LTS sample, we plan to use the CBCL parent- and teacher-report measures around attention to identify if any individuals were reported to be within the clinical range for attention problems. Additionally, we plan to use the executive function and impulsivity measures. We may investigate whether there are diverging patterns of brain development in those who had attention problems in childhood and whose symptoms worsened in adulthood versus those who had attention problems in childhood but whose symptoms did not change through adulthood versus those who did not have attention problems both from childhood to adulthood. This project is in early development.

ali.sloan.1@vanderbilt.edu

Sloan, A.F., Iacono, W.G., McGue, M., Wilson, S. (+ IBG Investigator team), Schaefer, J.D.

Tobacco use, a modifiable behavior, is associated with increased risk of psychotic symptoms. The proposed paper aims to test for evidence of a causal relationship between adolescent tobacco use and adult psychoticism using repeated, prospective self-report and interview assessments of adolescent tobacco use in five twin cohorts prospectively assessed over two decades. Analyses will compare twins from the same family to test whether the twin who uses more tobacco scores higher on our measure of psychoticism, on average, than their twin who reports less tobacco use. If associations between tobacco use and psychoticism are explained by shared familial confounds, this effect will be null. Significant within-twin-pair effects, on the other hand, would be consistent with, though insufficient for definitively establishing, causality. Tobacco use data was collected prospectively beginning in adolescence for all cohorts (ES, MF, ML, CTS, LTS); past-180-day frequency of tobacco use will be the primary measure of adolescent tobacco use exposure, and alternative measures might include lifetime tobacco use and age of first use. Twins were assessed for psychosis-proneness using the Psychoticism scale from the Personality Inventory for DSM-5 (PID-5) in adulthood. We plan to test for relationships between adolescent tobacco use and each psychoticism outcome in adulthood (total score and each subscale) across the pooled sample (i.e., combining ES, MF, ML, CTS, LTS cohorts). We then plan to test whether the relationship between tobacco use and psychoticism persists after controlling for shared familial risk factors and twin differences in non-target substance use (e.g., cannabis and alcohol use).

https://drive.google.com/open?id=1x6xGsmMNw4B9WypC3WLJcUMTpZV_OpuI

tessa.awald@colorado.edu

Addiction is often misunderstood, with many still viewing it as a personal failing rather than a complex issue influenced by genetics and environment. Studying how public opinions on drug use have changed over time, especially in response to scientific discoveries, can reveal how society shapes policies, treatment approaches, and stigma around addiction.
Understanding these shifts is not only fascinating but also important for promoting more informed, compassionate, and effective responses to substance dependence. Tessa proposes to use data on attitudes towards harm and other psychosocial variables to investigate how substance use (particularly nicotine, cannabis and opioids) has changed over time, comparing adolescent use in the CTS and CADD (first tested in 1998) compares to adolescent use in the COTWINS sample (tested in 2016). She will also examine attitudes towards use and patterns of use over time within the CTS and CADD samples.

n.richards097@utexas.edu

CATSLife, CATSLife1 and the ambulatory assessment

Elizabeth Muñoz; Anqing Zheng; Sally J. Wadsworth; Martin J. Sliwinski; Chandra A. Reynolds

While exposure to stressors has been widely examined in the literature in relation to cognitive function, stress severity– the extent to which stressors are perceived as threatening or overwhelming– has emerged as a critical determinant of cognitive function and change. However, the momentary effects of stress severity on everyday cognitive performance remain unclear. This study will first examine the momentary associations between stress severity and cognitive performance, and then further evaluate contextual factors, such as personal characteristics and daily activities, that may impact associations between stress and cognition. We will use ecological momentary assessment data from the Colorado Adoption/Twin Study of Lifespan behavioral development and cognitive aging. Participants (N = 440, Mage = 35.81, 57.7% women) completed momentary surveys and tasks assessing specific domains of cognition (i.e., working memory, executive functioning, speed, and episodic memory) via smartphones up to three times daily for 14 days. We plan to use multilevel modeling, accounting for momentary and between-person confounders, to examine daily associations. We will test new approaches for the treatment of time-structured data to model between and within-person effects (i.e., McNeish & Matta, 2020; Grimm et al., 2021). With these findings, we hope to highlight the importance of momentary context in understanding stress-related cognitive vulnerability.

Jessica.M.Ross@cuanschutz.edu

CADD, LTS, CTS, COTWINS

Will examine whether childhood/adolescent ADHD symptoms and cannabis use interact to predict cognitive function in adulthood. This will use my K23 data.

Not yet. Need to see if childhood and adolescent ADHD symptoms are available

kape8874@colorado.edu

Michael C. Stallings and Daniel E. Gustavson

The age-old competition between nature and nurture continues to be a crucial debate within behavioral genetics. The LTS dataset measures musical aptitude via three tests: Number of Notes–which is essentially identifying the number of notes used for a specific chord; Pitch Change–where the participant will hear two chords and need to report the trajectory of the note change; and last but not least, Melodic Memory–in which participants detect the altered note in a pair of otherwise identical melodies. While LTS is a longitudinal study, these measures were only recorded when participants were 12 years old. However, the adoption design allows further research in examining heritability. By analyzing sibling correlations across diverse family structures–including biological, adoptive, half-siblings, and step-siblings–and using z-scored parent-child comparisons to standardize scales, this study explores the extent to which musical aptitude is heritable versus environmentally shaped. Gene-by-environment (GxE) models will also be used to test whether genetic influences on musical aptitude varies by familial context, such as adoption status or sibling age spacing, offering insight into how nature and nurture jointly shape complex cognitive traits.

daniel.gustavson@colorado.edu

Lydia Rader, Naomi Friedman, Chandra Reynolds, Soo Rhee, Hermine Maes

Music engagement is a multidimensional human experience, playing roles in identity, skill mastery, and emotion regulation, yet its psychological functions and genetic architecture are still understudied. This study aims to identify latent constructs underlying musical engagement through exploratory and confirmatory factor analysis (EFA & CFA) of self-report items administered in COMN Paths. The items include measures of musical ability (e.g., rhythm, memory), musical identity and engagement, and music-based emotion regulation (e.g., using music to manage stress or loneliness). We will use the family structure of COMN Paths to uncover the heritability of these constructs and assess their intergenerational transmission. Furthermore, we will examine phenotypic relationships between latent music constructs and substance use experimentation and frequency and evaluate how individual differences (e.g., impulsivity, polygenic scores for alcohol consumption) may moderate these associations. This work will inform which facets of music engagement may serve as a protective or modifying factor in substance use risk. By characterizing the structure, heritability, and correlates of music-related traits, this research contributes to a more integrative understanding of how music functions in development, regulation, and resilience.

https://drive.google.com/open?id=1Ndx1nR-pqCHXan80HNq5uYb2An2nNlxy

CADD, LTS, CTS, CATSLife, COMN/COMN Paths

Arpana Agrawal; Ryan Bogdan; Melissa A. Cyders; Nisha C. Gottfredson O'Shea; Leslie A. Hulvershorn

Differential associations between distinct facets of impulsivity (i.e., lack of planning, sensation seeking, and negative/positive urgency), alcohol use, and alcohol use disorder (AUD) are thought to reflect etiological processes that differ with respect to stages of both alcohol involvement and lifespan development. Impulsive traits are frequently identified as among the strongest personality predictors of heavy alcohol use and related problems. However, our understanding of the longitudinal links between impulsive traits and alcohol involvement across development is generally limited to studies focusing on specific epochs (e.g., late adolescence). Given the potential etiological and clinical importance of developmental expression and changes in impulsive traits with respect to AUD, large-scale investigations of these links across a broader developmental span are needed. My current research (K01AA031724) is, in part, focused on characterizing associations between distinct impulsive personality traits and alcohol involvement across development through harmonization and longitudinal integrative data analysis of several extant datasets with assessments spanning childhood to middle adulthood. Using integrative data analysis to combine and analyze together longitudinal data from multiple independent studies can yield the statistical power to evaluate potential effects not observable in individual cohorts and may provide a more comprehensive understanding of this phenomenon that individual studies can offer. As part of this ongoing effort, this project will utilize relevant impulsivity and alcohol use/problems data from CADD assessments, data on year 16 assessments of impulsivity from CAP and LTS, and any relevant data available from the CATSLife 1 and COMN 1 grants. As a collaboration, these longitudinal data would be integrated alongside other acquired data from similar longitudinal studies (e.g., ABCD, Add Health, ALSPAC, COGA, NCANDA, Spit for Science, TRAILS) to generate a harmonized, large-scale dataset to be used in longitudinal analyses of bidirectional associations between changes in impulsive personality traits and progression of alcohol involvement (i.e., drinking milestones, frequency of typical and heavy drinking, and emergence of related consequences and problems).

https://drive.google.com/open?id=1khejY_-qShIk-DLgg5hIg7xYwMA7Wblt

saca5890@colorado.edu

Alaina Di Dio, Chandra Reynolds, Soo Rhee

The proposed study will examine phenotypic and etiologic associations between youth antisocial behavior and symptoms of antisocial personality disorder (ASPD) and psychopathy in adulthood. In the LTS and CAP samples, we plan to use the rule-breaking and aggression scales from the Achenbach instruments to model trajectories of antisocial behavior across childhood and adolescence (ages 7-16 years). Trajectories of rule-breaking and aggression will be identified using growth mixture modeling. Next, we will examine these trajectories as predictors of ASPD symptoms and psychopathy during emerging adulthood (age 23 years) in the LTS and CAP samples, as well as during midlife (~ages 30-50 years) in the CATSLife project. Lastly, we will examine the genetic and environmental contributions to the overlap between rule-breaking and aggression trajectories, respectively, and ASPD/psychopathy in adulthood.

https://drive.google.com/open?id=1XM5wxobgmKbYAmzabDJIeF24mBtkCV4X

saca5890@colorado.edu

Alaina Di Dio, Chandra Reynolds, Soo Rhee, Luke Hyde, Kelly Klump, Alex Burt

In the proposed study, we plan to examine genotype-environment interaction as a dynamic process that contributes to youth trajectories of antisocial behavior. In the LTS and CAP samples, we plan to use the rule-breaking (RB) and aggression (AGG) scales from the Achenbach instruments to examine the development of behavior problems throughout childhood and adolescence (ages 7-16 years) via growth curve modeling and mixture modeling. Subsequent analyses will leverage the LTS and CAP Census data on neighborhood characteristics to evaluate whether the genetic and environmental contributions to trajectories of antisocial behavior shift as a function of neighborhood disadvantage. We will also run these analyses using data on neighborhood disadvantage and youth RB and AGG from the Twin Study of Behavioral and Emotional Development in Children and the Michigan Twin Neurogenetics Study, within the Michigan State University Twin Registry.

https://drive.google.com/open?id=1JvuLxYVzZrdKqY7dhtpBDlJ36RifEZNp

linghua.jiang@colorado.edu

Daniel Gustavson, Lilia Geel, Soo Rhee, Chandra Reynolds

Although prior research has established robust links between negative parenting and children’s internalizing and externalizing problem behaviors across diverse cultural contexts (Pinquart, 2021; Vazsonyi et al., 2022), less is known about how these associations evolve over time, particularly across the transition from late childhood to adolescence. Moreover, few studies have employed adoption designs to explore how parenting and genetic background may interact across development using time-varying models. Drawing on data from the Colorado Adoption Project (CAP) and the Longitudinal Twin Study (LTS), the present study applies Time-Varying Effect Modeling (TVEM) to examine how the association between negative parenting and child problem behaviors changes from age 9 to age 16. In addition, we will investigate whether these time-varying associations differ between adopted and non-adopted youth.

https://drive.google.com/open?id=1ySxudmbbwp-ny4qz5OIn9TDxk02tzI4F

chandra.reynolds@colorado.edu

Anqing Zheng, Elizabeth Munoz, Robin P. Corley, Daniel E. Gustavson, Naomi P. Friedman, Robert Plomin, Sally J. Wadsworth

Cognitive dispersion—within-person inconsistency across tasks—may signal dementia risk in older adults, but its origins and relevance before midlife are unclear. We examined dispersion in established adulthood as a potential marker of cognitive reserve, and we distinguished it from general ability using heritability analyses.

https://drive.google.com/open?id=1g0PBHeUnPLJFeZ0-qFswoxVEdYnJqoAJ

chandra.reynolds@colorado.edu

Ryan Bruellman, Donald Evans, Andrew Smolen, Luke M. Evans

We compared DNA methylation in three tissues collected in 91 siblings and twins from the Colorado Adoption/Twin Study of Lifespan behavioral development and cognitive aging (CATSLife1): saliva, buffy coat (BC), and peripheral blood mononuclear cells (PBMC).

https://drive.google.com/open?id=190UuPQ2ibcjduKIAvQ6RsQtxtj1j783c

chandra.reynolds@colorado.edu

Daniel Gustavson, Naomi Friedman, Shandell Pahlen, Robin Corley, Matthew Keller, Sally Wadsworth

We examine cross-sectional and longitudinal associations of APOE, AD PGS, and EA PGSs with cognitive trajectories in individuals approaching midlife (30-45 years) from the ongoing Colorado Adoption/Twin Study of Lifespan behavioral development and cognitive aging study (CATSLife), with over 30 years of follow-up from parent studies (Colorado Adoption Project, Longitudinal Twin Study).

https://drive.google.com/open?id=1CP4K-AscDexY0R1JmYCoTAVJTAms3J9k

chandra.reynolds@colorado.edu

Andrew Smolen, Christopher D. Link, Donald Evans, Luke Evans, Elizabeth Munoz, Sally J. Wadsworth

Neurofilament light chain (NfL) is a biomarker indexing axonal integrity where small NfL variations may be associated with cognitive performance in early adulthood and high values associated with neurodegenerative disorders such as Alzheimer’s disease. However, associations of NfL and related biomarkers of neurodegneration with performance versus variability in performance are unclear. We evaluate inconsistency in trial-level performance on three tasks tapping speed or spatial reasoning and dispersion across a battery of verbal, spatial, speed, and episodic memory tasks in participants from the Colorado Adoption/Twin Study of Lifespan behavioral development and cognitive aging (CATSLife1).

https://drive.google.com/open?id=12EJGEYJeZpSBaE_ZylA2JYs-0TZKrF6b

Chandra Reynolds, Ilana Bennett, Anqing Zheng, Dan Gustavson

Fitness tracker devices offer physical activity-related behavior across time but also allow for a detailed understanding in the temporal aspect. Through the data available from fitness trackers including metabolic equivalent (MET) scores, heart rate data, and step data, ascertaining the when and how much questions relevant to the physical activity guidelines can be gleaned. Previous insight into the temporal patterns of exercise have found positive health outcomes for those participating in regular physical activity, with those loading physical activity on weekends (i.e. weekend warriors) seeing benefits. However, no studies to date have utilized twin pairs in parsing apart temporal exercise trends with health outcomes.
This study will utilize fitness tracker data collected during the second wave of CATSLife. Such data includes MET score data (tracked by minute), heart rate (tracked by minute), and step count (tracked by minute). To uncover temporal trends across these various measures functional principal component analysis (fPCA) will be used due to its unique ability to uncover and identify temporal trends and has been utilized as such with fitness tracker data. Running fPCA across participants will provide a score which will classify them into distinct groups based on their physical activity behavior. Health outcomes (i.e. BMI, lipids, heart metrics, etc.) across these groups will then be compared to determine whether health outcomes significantly vary based on temporal patterns of physical activity. Twin pairs will also be compared to further parse apart genetic and environmental associations between temporal physical activity patterns and health outcomes.

anqing.zheng@colorado.edu

Ryan Bruellman, Sarah Carroll, Claudia Paniagua-Ugarte, Joey White, Robin Corley, Luke Evans, Naomi Friedman, Daniel Gustavson, Andrew Smolen, Sally Wadsworth, Matthew Keller, Chandra Reynolds

Examine the association between BCS and cognition, frailty, and brain age delta with CATSLife sample

https://drive.google.com/open?id=1BP3BUuN3eAoEFHhYCi_feQrTpBSGyDKF