Problem-solving Behavior Research Articles

Achievement emotions in kindergarten: the association of solution accuracy with discrete joy, sadness, and surprise

Children experience a variety of emotions in achievement settings. Yet, mathematics-related emotions other than anxiety are understudied, especially for young children entering primary school. The current study reports the prevalence and intensity of six basic, discrete achievement emotions (joy/happiness, sadness, surprise, anger, fear, and disgust) expressed on the faces of 15 kindergarten-aged children as they solved increasingly complex arithmetic story problems in a 3-month teaching experiment. We also examine how the extent to which the expressed emotions influenced arithmetic accuracy at the end of an instructional session at the beginning, middle, and end of the teaching experiment. Through the application of FaceReader9, the three most intensely expressed emotions at the launch of the instructional sessions were happiness/joy, sadness, and surprise. Using functional regressions, these expressed achievement emotions predicted arithmetic accuracy at the end of the instructional session. However, when the effect of session over time was added to the model, the relationship between happiness/joy and accuracy, as well as sadness and accuracy, became non-significant. In contrast, the relationship between surprise and accuracy remained significant. We discuss potential explanations for these patterns of significance and non-significance. This study serves as a critical first step in clarifying how emotions contribute to problem-solving behavior as we grapple with how to respond to the sometimes intense, but always present emotions of young learners in ways that are affirming, as well as mathematically productive and generative.

Understanding the Differences in an AI-Based Creativity Support Tool Between Creativity Types in Fashion Design

As the perspective on creativity shifts to “how it is expressed,” research has aimed to categorize it by problem-solving style. Since harnessing individual creative traits can positively impact creative performance, there has been an emphasis on designing computer systems that are tailored to personal problem-solving behaviors. AI-CST has opened up the potential to facilitate such customization. In this work, we consider two types of creativity—adaptors and innovators—based on problem-solving styles, and investigate AI-CST designs that both types could flexibly use according to the fashion design process. We identified two main AI-CST functions—determining design direction and receiving design inspiration—of the fashion design process, and developed CoCoStyle to map these functions. Through a user study with 30 fashion professionals (15 adaptors and 15 innovators), we found significant differences between the two groups from survey responses, system usage logs, and interviews. Based on the results, we discuss the theoretical and practical implications of AI and AI-CST where creativity is essential.

A sparse latent class model incorporating response times.

Diagnostic models (DM) have been widely used to classify respondents' latent attributes in cognitive and non-cognitive assessments. The integration of response times (RTs) with DM presents additional evidence to understand respondents' problem-solving behaviours. While recent research has explored using sparse latent class models (SLCM) to infer the latent structure of items based on item responses, the incorporation of RT data within these models remains underexplored. This study extends the SLCM framework to include RT, relaxing the conditional independence assumption between RT and latent attributes given individual speed. This adaptation provides a more flexible framework for jointly modelling RT and item responses. While the proposed model holds promise for applications in educational assessment, this study applied the model to the Fisher Temperament Inventory, yielding findings that provide a novel perspective on utilizing DM with RT in personality assessments. Additionally, a Gibbs sampling algorithm is proposed for parameter estimation. Results from Monte Carlo simulations demonstrate the algorithm's accuracy and efficiency.

Assessing the Coping Strategies of Healthcare Workers During COVID-19 in Jaipur, Rajasthan: An Empirical Analysis

This study explores the coping strategies adopted by healthcare workers (HCWs) during the COVID-19 pandemic and their relationship with stress levels. A balanced representation of doctors and nurses participated, with a slightly higher proportion of males (60%) and most respondents aged 25–44, reflecting a workforce likely to experience significant occupational stress. Among coping strategies, problem-focused approaches, such as planning and addressing stressors, were the most commonly utilized (54% occasionally), though only 9% of HCWs employed them daily due to the high-pressure environment. Emotion-focused strategies, including seeking emotional support, were moderately used (39%) and played a key role in stress management, but only 28% used them daily. Avoidant-focused coping strategies, such as denial and self-distraction, emerged as the most frequently employed daily mechanism (30%), offering short-term relief but raising concerns about long-term stress implications. Regression analysis revealed significant relationships between stress levels and coping mechanisms. Avoidant-focused strategies exhibited the strongest association (R² = 0.812), followed by emotion-focused (R² = 0.765) and problem-focused coping (R² = 0.681). The findings highlight the critical need for interventions promoting sustainable stress management strategies. Enhancing problem-solving behaviors, emotional regulation, and providing organizational support are essential to improve HCWs' well-being and healthcare system resilience during crises.

A language model of problem solving in humans and macaque monkeys.

Human intelligence is characterized by the remarkable ability to solve complex problems by planning a sequence of actions that takes us from an initial state to a desired goal state. Quantifying and comparing problem-solving capabilities across species and finding their evolutionary roots are critical for understanding how the brain carries out this intricate process. We introduce the Language of Problem Solving (LoPS) model as a novel quantitative framework that investigates the structure of problem-solving behavior through a language model. We applied the model to an adapted classic Pac-Man game as a cross-species behavioral paradigm to test both humans and macaque monkeys. The LoPS model extracted the latent structure, or grammar, embedded in the agents' gameplay, revealing the non-Markovian temporal dependency structure of their problem-solving behavior and the hierarchical structures of problem solving in both species. The complexity of LoPS grammar correlated with individuals' game performance and reflected the difference in problem-solving capacity between humans and monkeys. Both species evolved their LoPS grammars during learning, progressing from simpler to more complex ones, suggesting that the structure of problem solving is not fixed but evolves to support more sophisticated and efficient problem solving. Our study provides insights into how humans and monkeys break down problem solving into compositional units and navigate complex tasks, deepening our understanding of human intelligence and its evolution and establishing a foundation for future investigations of the neural mechanisms of problem solving.

The Critical Role of Tangible User Interfaces in Human Computer Interaction

Human-Computer Interaction has evolved from command-line, then to graphical, up to a tangible user interface (TUI). TUIs represent a new paradigm in incorporating physical objects within the digital environment in order to offer users richer, more natural, and intuitive interaction means. This paper reviews the applications of TUIs within cognitive ergonomics, education, and industry, with special emphasis on the potential effects that TUI might have in reducing cognitive load and improving retention and enhancing problem-solving behavior. It covers various case studies on cognitive benefits at TUI, frameworks of distributed and embodied cognition, scalability and accessibility issues, ways to reduce technical obstacles along with users' reluctance, and how TUI is being merged with IoT. The authors also discuss how TUI will see huge improvements in terms of networking and control in intelligent environments. From the above, though TUIs promise great benefits related to the conventional GUIs, full utilization in different applications calls for addressing cost, adaptability, and inclusivity for wide usage.

Identifying quality responses using an analysis of response times: the RTcutoff function in R.

The present study aims to develop an R function to develop and visualize thresholds that describe the response time of individuals concerning their sample. The function utilizes the cumulative proportion correct (CUMP) approach, to estimate item-specific time threshold, which originated in the work of Guo and his colleagues. Besides the CUMP approach, the present function presents response time profiles on a measure using the mean of the sample and + 1SD times so that it can discern between thoughtful engagement and processing with an item (termed problem-solving behavior) and rapid responding, guessing, and disengagement with the test. The advantage of the CUMP model utilized here is that it simultaneously engages both response time and response correctness to establish thresholds that differentiate engaged from disengaged participants. Given data on a measure of reading comprehension for students in Saudi Arabia (n = 494) using the Progress in International Reading Literacy Study (PIRLS 2021) international assessment, high and low-achieving individuals that engaged in different behavior patterns were identified and plotted against their sample. Results pointed to the importance and necessity of the RTcutoff function to identify variable forms of engagement that have implications for person-score validity but also have implications for test validity and the need to increase measurement precisio.

COVID-19 Pandemic Coping, Social Support, and Emotional Health in American Indian and Alaska Native Peoples

The COVID-19 pandemic has placed a burden on the health of many people, including significant disparities in American Indian and Alaska Native communities. This study examines the associations between coping behaviors, social support, and emotional health among American Indian and Alaska Native peoples during the COVID-19 pandemic. This cross-sectional study included survey data collected from November 2021 to May 2022 from American Indian and Alaska Native adults aged 18 years or older without dementia or other serious cognitive impairments who were seen at 6 urban health organizations primarily in urban settings (in New Mexico, Alaska, Colorado, Minnesota, Utah, and Kansas) in the year prior to the survey. Exposures of interest included avoidant and direct problem-solving coping behaviors and functional and emotional social support. The study outcome was self-reported change in emotional health since COVID-19 pandemic onset. Poisson regression was used to model adjusted multivariate associations. Data were weighted to account for age, nonresponse, and disproportionate representation by clinic population. A total of 1164 participants were included in the analysis, with a mean (SD) age of 42.5 (13.4) years; 830 (61%, weighted sample percentage) were female. Since COVID-19 pandemic onset, 465 patients (39% weighted) reported worsened emotional health. Problem-solving coping mean (SD) utilization score was 2.5 (0.5), avoidant coping mean (SD) utilization score was 2.3 (0.5), mean (SD) functional social support score was 11.4 (2.9), and 219 participants (18% weighted) reported that emotional support was always available. Using problem-solving coping skills was associated with better emotional health (adjusted prevalence ratio [APR], 0.66 [95% CI, 0.54-0.81] for highest vs lowest tertile), as was always (vs never or rarely) getting emotional support (APR, 0.40; 95% CI, 0.30-0.55) and having more functional support (APR, 0.90 [95% CI, 0.87-0.92] per 1-unit increase in functional social support). In examination of psychological resilience potentially modifying primary exposure associations, no interactions were statistically significant. In this cross-sectional study of urban American Indian and Alaska Native peoples, problem-solving coping skills and more social support were associated with better emotional health during the COVID-19 pandemic. These findings can be used to identify strengths-based approaches to support community emotional health during social upheavals.

Feasibility and Acceptability of a Self-Guided Digital Family Skills Management Intervention for Children Newly Diagnosed With Type 1 Diabetes: Pilot Randomized Controlled Trial.

Family dynamics play an important role in determining the glycemic outcomes of type 1 diabetes (T1D) in children. The time interval immediately following T1D diagnosis is particularly stressful for families, and interventions to support families in adjusting their family practices to support adjustment to and management of T1D in the months following diagnosis may improve glycemic outcomes. Self-guided digital interventions offer a sustainable model for interventions to support caregivers in learning evidence-based family management skills for adjustment to and management of T1D. We hypothesized that a self-guided, web-based, family skills management program (addressing caregiver social support as well as family problem-solving, communication, and supportive behavior change strategies) initiated at the time of T1D diagnosis would improve glycemic outcomes in children with T1D. In this study, we report on the feasibility and acceptability of this program. We prospectively evaluated a sample of 37 children newly diagnosed with T1D recruited from a pediatric endocrinology clinic. Parent participants were asked to complete web-based modules addressing social support, family problem-solving, communication, and supportive behavior change strategies. Module completion was analyzed for percentage completion, patterns of completion, and differences in completion rates by coparenting status. Qualitative open-ended feedback was collected at the completion of each module. A total of 31 (84%) of the 37 participants initiated the web-based program. Of those 31 participants, 25 (81%) completed some content and 15 (48%) completed all 5 modules. Completion rates were higher when coparenting partners engaged in the intervention together (P=.04). Of the 18 participants given a choice about the spacing of content delivery, 15 (83%) chose to have all sessions delivered at once and 3 (17%) chose to space sessions out at 2-week intervals. Qualitative feedback supported the acceptability of the program for delivery soon after T1D diagnosis. Families reported on positive benefits, including requesting future access to the program and describing helpful changes in personal or family processes for managing T1D. In this study, we found that a self-guided digital family support intervention initiated at the time of a child's T1D diagnosis was largely feasible and acceptable. Overall, rates of participation and module completion were similar to or higher than other self-guided digital prevention interventions for mental and physical health outcomes. Self-guided digital programs addressing family management skills may help prevent challenges common with T1D management and can decrease cost, increase access, and add flexibility compared to traditional interventions. ClinicalTrials.gov NCT03720912; https://clinicaltrials.gov/study/NCT03720912.

Catatonia and neurocognitive disorder in the context of schizophrenia spectrum disorders

Background. Currently, researchers are aiming to establish the relationship between neurocognitive disorder and other clinical manifestations of endogenous diseases and functional indicators of the patient’s adaptation to social life. According to some data, the development of cognitive disorders is associated with the severity of catatonic disorders in the clinical picture of various diseases, in particular the presence of catatonic symptoms determines the worst indicators of cognitive functioning in the case of schizophrenia.The aim. To study the clinical and psychopathological features of catatonic syndrome in connection with schizophrenia spectrum disorders, as well as its connection with neurocognitive disorder manifestations among patients with endogenous disorders.Methods. From September 2022 to March 2024, we conducted a prospective, multicenter study of patients (n = 69) suffering from paroxysmal schizophrenia with primary catatonic symptoms in the structure of attacks. Mental status, the severity of catatonic symptoms using Bush – Francis Catatonia Rating Scale, neurocognitive profile using a number of validated scales (Montreal Cognitive Scale, Brief Assessment of Cognition in Schizophrenia, Frontal Assessment Battery) were assessed.Results. Neurocognitive disorder was revealed in all cases studied. It has been noted that, depending on the presence/absence of affective symptoms, various domains of the neurocognitive profile are affected. Thus, in patients without severe affective symptoms, auditory and verbal memory, attention, processing speed, and motor skills are impaired. In patients with pronounced affective disorders, processing speed, motor skills, and problem-solving behavior are affected.Conclusion. Based on the results of processing the material, it was revealed that the most pronounced neurocognitive disorders were found in case of the manifestation of paroxysmal paranoid schizophrenia in male patients with a relatively early onset of the disease and hypokinetic catatonic disorders. In the future, we are planning to compare these results with the results of control group of patients with endogenous diseases and without catatonia.

Predicting longitudinal trajectories of adaptive couple processes in families of autistic children: Applying the Vulnerability Stress Adaptation model

BackgroundMany couples raising an autistic child report long-standing and thriving couple relationships, even within the context of elevated parenting stress. Little is known however, about the factors that promote or hinder adaptive couple processes or how these processes may shift over time. Drawing on the Vulnerability Stress Adaptation (VSA) model, the present study examined adaptive couple problem-solving processes across three years and the extent to which these processes were predicted by parent broader autism phenotype (BAP) and level of parenting stress in couples raising an autistic child. MethodIn total, 189 couples (378 parents) raising an autistic child (aged 5–12 years) completed questionnaires assessing parent BAP and level of parenting stress. Couples also engaged in observed problem-solving interactions at four time points spanning three years. ResultsThe use of adaptive couple problem-solving behaviors (Solution Focused Reciprocity and Positive Teamwork) declined linearly across time. Father BAP predicted greater decline in these adaptive processes (Solution Focused Reciprocity and Positive Teamwork), particularly when mother-reported parenting stress was low. No main effects of parenting stress on change in adaptive couple problem-solving behaviors emerged. ConclusionsFindings highlight the utility of the VSA Model for understanding the short-term longitudinal course of problem-solving behaviors in couples raising an autistic school-aged child. This study can inform efforts to promote healthy and long-lasting couple relationships in this population.

Productive Problem-Solving Behaviors of Students with Learning Disabilities

ABSTRACT The purpose of this study was to explore the problem-solving behaviors of middle-school students with learning disabilities (SLD). Think-aloud interviews were performed with 20 seventh- and eighth-grade students who had learning disabilities to observe their behaviors while solving mathematical word problems (i.e., behaviors and patterns of behaviors). Themes emerged from qualitative analysis indicating that the students engaged in problem-solving behaviors, as well as common patterns of behaviors. There were two consistent behaviors: (a) verbalizing a plan before moving forward on a problem and (b) reengaging after hitting an obstacle in the problem. Data also revealed common patterns of behavior that suggested participants were (a) sensemaking and (b) reflecting throughout the problem-solving process. Findings suggest that these participants, middle-grade students with learning disabilities (SLD), were engaging in productive problem-solving behaviors and persevering throughout the process. Future studies are discussed that maintain an asset-based view on students with learning disabilities.

“Although Burdened, Do We Need to Do More?” Street-Level Bureaucrats’ Organizational Citizenship Behaviors in Poverty Alleviation Policy Implementation

Drawing on attentional resource allocation theory and transaction cost theory, this study links street-level bureaucrats’ problem-solving organizational citizenship behaviors with perceived administrative burden. Based on 28 in-depth interviews and 657 survey responses from street-level bureaucrats involved in China’s Targeted Poverty Alleviation Policy during 2019 and 2020, the study found positive relationships between problem-solving organizational citizenship behaviors, resource deficiency, bureaucratic control, and perceived administrative burden. Policymakers should be cautious when using control tools and avoid exploiting street-level bureaucrats’ commitment and compassion for clients to ensure effectively policy implementation and reduce administrative burdens on dedicated public servants.

Autonomous 4? Learners’ Collaborative Problem Solving in Chamber Music Rehearsals With and Without a Coach

This investigation explored collaborative problem solving with and without a coach by analyzing problem-solving (PS) and problem-solving-prompting (PSP) behaviors by group members and coaches in six total coached rehearsals with three collegiate string quartets, and six autonomous rehearsals with the same string quartets. Teachers’ and learners’ PS and PSP behaviors were compared in three settings: autonomous chamber music rehearsals, chamber music rehearsals with a coach, and one-to-one instruction. Coaches and students performed the same problem-solving component behaviors and most of the same PSP behaviors as observed in one-to-one instruction and in rehearsals of a professional chamber ensemble. As in one-to-one instruction, coaches could engage learners in problem solving, and in this setting invite them to collaborate with one another, by asking questions and varying the specificity of feedback and directives. Coaches’ Attention Directives preceded the most learner problem solving and collaborative engagement. During autonomous rehearsals learners often switched leadership roles fluidly, often prompting one another to problem solve. Implications for a pedagogy of coaching and inclusion of small ensembles within music education are discussed.

Accuracy and effectiveness of an orchestration tool on instructors’ interventions and groups’ collaboration

This paper presents the development of a novel orchestration tool that predicts collaborative problem-solving (CPS) behaviors of undergraduate engineering groups and investigates the use of that tool by instructors. We explore the impact of receiving real-time, machine-learning, model-based prompts on 1) instructors’ orchestration strategies, which are strategies instructors use to manage and facilitate collaborative activities, and 2) groups’ participation, including how groups are engaged in CPS activities. The orchestration tool is a dashboard that notifies instructors of—and advises them on—monitoring and intervening with groups who may need collaborative support and guidance. We describe the accuracy of the models in predicting CPS behaviors and of instructors in identifying these behaviors in the classroom. We then describe how real-time prompts from models can affect instructors’ orchestration strategies and students’ participation. Our findings show that there is variability in the accuracy of our machine learning models and that instructors are better at identifying predictive behaviors as compared to the models. Instructors in this context engaged in orchestration strategies, like monitoring and probing when using the orchestration tool, and groups of students were largely talking while on-task across classes. We triangulate across data sources to examine the effectiveness of the orchestration tool in the classroom and share pedagogical and technical implications for the field.

Psychometric Testing of the Korean Version of the Self-Care of Coronary Heart Disease Inventory Version 3

PurposeThis methodological study evaluated the psychometric properties of the Self-Care of Coronary Heart Disease Inventory version 3 (SC-CHDI v3) in a Korean context. MethodsThe SC-CHDI v3 was translated into Korean following a rigorous translation process. Participants were 452 patients who had experienced coronary heart disease (CHD), all recruited from a tertiary hospital in Korea. Exploratory and confirmatory factor analyses were performed to test construct validity. Concurrent validity was examined by correlating scores from the Korean version of the SC-CHDI v3 with those from the Cardiac Self-Efficacy Scale. Internal consistency was analyzed using Cronbach’s alpha and McDonald’s omega. ResultsThe Korean version of the SC-CHDI v3 consists of 21 items, excluding two from the original instrument. The self-care maintenance subscale identified a two-factor structure: “treatment adherence” and “health-promoting behaviors.” The goodness-of-fit indices were satisfied: χ2 = 18.19, p = .110, comparative fit index (CFI) = .97, Tucker-Lewis Index (TLI) = .95, and standardized root mean square residual (SRMR) = .04. The self-care monitoring subscale consisted of a one-dimensional structure (“monitoring behaviors”) and the goodness-of-fit indices were satisfied: χ2 = 19.19, p = .059, CFI = .99, TLI = .99, and SRMR = .04. The self-care management subscales had a two-factor structure of “consulting behaviors” and “problem-solving behaviors.” The goodness-of-fit indices were satisfied: χ2 = 16.44, p = .037, CFI = .99, TLI = .98, and SRMR = .03. Scores from the Cardiac Self-Efficacy Scale showed a positive correlation with the Korean version of SC-CHDI v3 subscales. Reliability estimates were ≥ .80 for all subscales except for the self-care maintenance subscale. ConclusionsThe Korean version of the SC-CHDI v3 consists of 21 items in 3 subscales and is a valid and reliable instrument. Therefore, healthcare providers can effectively utilize it to assess the self-care levels of patients with CHD.

Developing Tenth-Grade Students' Problem-Solving Skills in Chemistry through Deductive Learning with the DAPIC Approach

The authentic science classroom requires students to think and solve problems through hands-on activity. This research aims to enhance problem-solving skills in chemistry of grade 10 students through deductive learning with the DAPIC approach, with a passing criterion of 70% or higher of the total score. This research employed action research. The target group consists of 22 students from the model school of Mahasarakham University in the second semester of the academic year 2023. The data collection instruments included nine lesson plans, three sets of essay-type problem-solving tests (each comprising three questions), student interview questionnaires regarding their opinions on learning management, and observational data on problem-solving behavior in chemistry. The research findings indicated that the target group improved their problem-solving skills after learning through deductive learning with the DAPIC approach. In the first cycle, 50% of students (11 out of 22) met the passing criterion of 70%. In the second cycle, 81.82% of students (18 out of 22) passed, and in the third cycle, 95.45% of students (21 out of 22) achieved the passing criterion. It can be concluded that deductive learning with the DAPIC approach can also enhance students' problem-solving in the science classroom.

How story problems strengthen arithmetic problem-solving strategy sophistication: Evidence from a learning trajectory teaching experiment in kindergarten

BackgroundThe sophistication of young children's arithmetic problem-solving strategies can be influenced through experience and instructional intervention. One potential pathway is through encountering story problems where the location of the unknown quantity varies. AimsThe goal of the present study is to characterize how arithmetic problem-solving strategy sophistication can evolve through opportunities to solve story problems. SampleWe used microgenetic principles to guide the coding of arithmetic problem-solving behavior (8843 attempts) across three timescales (time within-session, attempt to solve, and between sessions) for nine story problem structures (N = 40, 19 girls). Data come from a teaching experiment conducted in a Mountain West US state in Spring 2018. MethodsWe employed a Bayesian hierarchical ordinal regression with a nine-level response variable. The model contained fixed effects for session, attempt, story problem structure; a smooth time within session effect; and random effects for student, instructor, and equation. ResultsOur analysis indicates which transitions from less to more sophisticated strategies are better supported by additional attempts to solve the same problem vs. additional instructional sessions. Strategy sophistication also varied by the location of the unknown quantity (result unknown, find difference, start unknown), but not operation (join, separate, part-whole). ConclusionsIf confirmed by other studies, including experiments, what teachers offer children in terms of learning opportunities (more attempts within the same problem or more problems across work sessions) should vary based on the transition they are making.

The Role of Social Problem-Solving and Prosocial Behavior in First Aid Willingness among Adolescents in Hungary.

First aid is a key factor in preventing further deterioration in an accident, saving lives, or improving treatment in emergencies. However, the reasons behind the willingness to provide first aid are still unclear. Therefore, this study aimed to investigate the role of social problem-solving and prosocial behavior in the dimension of first aid willingness. Self-administered questionnaires were used to evaluate the dimensions of first aid willingness (first aid willingness for peers, first aid willingness for strangers, knowledge, and negative emotions), social problem solving (positive problem orientation, negative problem orientation, rational problem solving, avoidance style, and impulsivity/carelessness) and prosocial behavior. A total of 497 school-aged students between the ages of 12 and 15 years (201 boys and 296 girls) participated in this study. Our results showed that positive problem orientation (p < 0.05) and rational problem solving (p < 0.001) are significant factors in determining first aid willingness for both peers and strangers. On the other hand, avoidance style orientation (p < 0.05) has a negative influence on the willingness to provide first aid to peers and strangers. Negative problem orientation (p < 0.001) only predicted negative emotions related to emergencies. Furthermore, prosocial behavior (p < 0.001) was more strongly associated with first aid willingness than social problem solving. Our study suggested that improving social competence could be a key factor in increasing first aid provision in real time, which could save lives in emergencies.

Predicting problem-solving success in an office simulation applying N-grams and a random forest to behavioral process data

Predicting students' problem-solving success in computer-based simulations at an early stage allows adaptive educational systems to provide learners with personalized support. In this paper, we predict students' problem-solving success by applying a machine-learning model, the random forest, to produce a binary classification (more vs. less successful students). During a business-related problem scenario that lasted 55 min, early behavioral data (during the first 5, 10, and 20 min) such as mouse clicks and keyboard strokes (approx. 29,800 early-window clickstreams and keystrokes during the first 20 min) of 234 trainees were recorded, mirroring the students' problem-solving behavior. We used the n-gram sequence mining technique, which was originally introduced within the emerging disciplines of natural language processing, text mining, and machine learning and has proven to be effective, particularly in the examination of online behavior. We trained the random forest model with training datasets that included all features (bigrams), as well as selected features (the most predictable bigrams explaining inter-group differences). Our results show that early predictions based on the first 10 and 20 min contained sufficient information to accurately predict problem-solving success, while predictions that are too early (based on the first 5 min) do not. As the size of the initial time window expanded, the classification performance improved. Moreover, the selection of the most predictable features improved the models' performance for all three time intervals. The model that was trained with only selected robust features that occurred in the first 20 min achieved the highest ROC AUC score of almost 0.70. This result falls within the range of accuracy scores observed in similar studies. From the instructor's perspective, predictions help in the early identification of weak students and can provide them with personalized learning prompts. For more successful students, tasks can be enriched adaptively.