Computer-generated Feedback Research Articles

The impact of weight self-stigma on weight-loss treatment engagement and outcome.

Individuals with overweight or obesity often endure significant weight-based prejudice and discrimination in various settings. Experiencing weight-related stigma is linked to many adverse psychosocial outcomes. Weight self-stigma is when an individual internalizes and identifies with negative attributes ascribed to people with larger bodies and has self-devaluing thoughts because of their weight and is associated with poorer health outcomes. This study explored how weight self-stigma may impact weight management efforts and outcomes for adults participating in an onlight weight-loss intervention. 508 adults (86.2% female, 84.6% White) with overweight or obesity participated in an asynchronous 12-week online weight-loss intervention with computer-generated feedback. Weight and weight self-stigma were measured at baseline and 3 months later. Thirty-one point five percent of the sample reported high levels of stigma, which was associated with greater program dropout than those who did not report high stigma (32.5% vs. 21.6%). Program completers reporting high self-stigma showed better treatment engagement (77.0% vs. 69.7% lessons viewed) and weight loss (M = -6.31% vs. -5.08%); these differences were not observed when using intent-to-treat assumptions. When analyzed as a continuous variable, weight self-stigma showed no association with treatment engagement and outcome. These findings highlight the complexity of understanding how self-stigma affects treatment engagement and outcome in behavioral weight loss and the need for more targeted research in this understudied area. Results suggest that weight self-stigma plays a role in weight management during an online weight-loss intervention, affecting engagement and outcomes.

From process to product: writing engagement and performance of EFL learners under computer-generated feedback instruction

Artificial intelligence (AI) technology is gradually penetrating the domain of education, opening up many possibilities for teaching and learning. Many educators, faced with the burden of commenting on substantial student essays, have introduced automated writing evaluation (AWE) into second language (L2) writing considering its affordance of immediate scores and diagnostic information. However, students’ processing strategies and perceptions of such computer-generated feedback and its impact on student writing quality, particularly as mediated by language proficiency, remain under-explored. This study examines the impact of Pigai, a Chinese AWE system, on revision processes and writing products of 42 English as a foreign language (EFL) learners with varying language levels by analyzing feedback points, feedback uptake, text quality in complexity, accuracy, and fluency (CAF), and perceptions. The findings are as follows. When confronted with AWE instruction, the majority of student work focuses on correcting errors, but higher-level students exhibit an emphasis on language improvement beyond the surface level compared to lower-level students. According to CAF measures, automated feedback exerts greater effects on accuracy for unskilled learners and lexical complexity for skilled learners in the development of interlanguage. Learner profiles and perceptions of students at different levels are explored along four dimensions: writing quality, cognitive engagement, behavioral engagement, and affective engagement. Finally, the potential issues of such technology-based writing instruction are pointed out.

The impact of a feedback intervention on university students’ second language writing feedback literacy

ABSTRACT This study evaluated the effect of a combined feedback activity (peer feedback, computer-generated feedback and teacher feedback) on students’ second language (L2) writing feedback literacy. One hundred and eighty-two Chinese university students participated in this research. Findings revealed that the intervention significantly improved students’ literacy in Appreciating Feedback, Acknowledging Different Feedback Sources and Managing Affect, but not Making Judgements and Taking Action. L2 proficiency levels affected the literacy development. Low-proficiency students’ feedback literacy did not change significantly. Middle-proficiency students improved significantly in Appreciating Feedback, Acknowledging Different Feedback Sources, Managing Affect, and Taking Action. High-proficiency students only improved significantly in Appreciating Feedback. Findings further reveal different degrees of difficulty for students to improve feedback literacy along its five dimensions. This study bears implications for developing students’ feedback literacy in L2 writing and in other disciplinary areas, particularly regarding how teachers could use multiple feedback sources and address students’ varied proficiency levels.

Artificial Intelligence-Powered Clinical Decision Support and Simulation Platform for Radiology Trainee Education.

Technological tools can redesign traditional approaches to radiology education, for example, with simulation cases and via computer-generated feedback. In this study, we investigated the use of an AI-powered, Bayesian inference-based clinical decision support (CDS) software to provide automated “real-time” feedback to trainees during interpretation of clinical and simulation brain MRI examinations. Radiology trainees participated in sessions in which they interpreted 3 brain MRIs: two cases from a routine clinical worklist (one without and one with CDS) and a teaching file-based simulation case with CDS. The CDS software required trainees to input imaging features and differential diagnoses, after which inferred diagnoses were displayed, and the case was reviewed with an attending neuroradiologist. An observer timed each case, including time spent on education, and trainees completed a survey rating their confidence in their findings and the educational value of the case. Ten trainees reviewed 75 brain MRI examinations during 25 reading sessions. Trainees had slightly lower confidence in their findings and diagnosis and rated the educational value slightly higher for simulation cases with CDS compared to clinical cases without CDS (p < 0.05). There were no significant differences in ratings of clinical cases with or without CDS. No differences in overall timing were found among the reading scenarios. Simulation cases with “CDS-provided feedback” may improve the educational value of interpreting imaging studies at a workstation without adding additional time. Further investigation will help drive innovation in trainee education, which may be particularly relevant in this era of increasing remote work and asynchronous attending review.

The Role of Sex and Age in Moderating the Outcome of In-Person and Computer-Based Brief Alcohol Interventions at General Hospitals: Reanalysis of a Brief Intervention Study

Introduction: The aim of this study was to test whether brief alcohol interventions at general hospitals work equally well for males and females and across age-groups. Methods: The current study includes a reanalysis of data reported in the PECO study (testing delivery channels of individualized motivationally tailored alcohol interventions among general hospital patients: in PErson vs. COmputer-based) and is therefore of exploratory nature. At-risk drinking general hospital patients aged 18–64 years (N = 961) were randomized to in-person counseling, computer-generated individualized feedback letters, or assessment only. Both interventions were delivered on the ward and 1 and 3 months later. Follow-ups were conducted at months 6, 12, 18, and 24. The outcome was grams of alcohol/day. Study group × sex and study group × age interactions were tested as predictors of change in grams of alcohol/day over 24 months in latent growth models. If rescaled likelihood ratio tests indicated improved model fit due to the inclusion of interactions, moderator level-specific net changes were calculated. Results: Model fit was not significantly improved due to the inclusion of interaction terms between study group and sex (χ²[6] = 5.9, p = 0.439) or age (χ²[6] = 5.5, p = 0.485). Discussion: Both in-person counseling and computer-generated feedback letters may work equally well among males and females as well as among different age-groups. Therefore, widespread delivery of brief alcohol interventions at general hospitals may be unlikely to widen sex and age inequalities in alcohol-related harm.

Proactive automatised lifestyle intervention (PAL) in general hospital patients: study protocol of a single-group trial

IntroductionThe co-occurrence of health risk behaviours (HRBs, ie, tobacco smoking, at-risk alcohol use, insufficient physical activity and unhealthy diet) increases the risks of cancer, other chronic diseases and mortality more...

Feedback That Leads to Improvement in Student Essays: Testing the Hypothesis that “Where to Next” Feedback is Most Powerful

Feedback is powerful but variable. This study investigates which forms of feedback are more predictive of improvement to students’ essays, using Turnitin Feedback Studio–a computer augmented system to capture teacher and computer-generated feedback comments. The study used a sample of 3,204 high school and university students who submitted their essays, received feedback comments, and then resubmitted for final grading. The major finding was the importance of “where to next” feedback which led to the greatest gains from the first to the final submission. There is support for the worthwhileness of computer moderated feedback systems that include both teacher- and computer-generated feedback.

L2 Writers’ Processing of Teacher vs. Computer-generated Feedback

Writing is thought as the most complicated skill in second language acquisition; therefore, L2 researchers have always been in pursuit of discovering an effective approach to improve it. One of the most debated ways is feedback which has a key role in improving the quality of writing. Much of the previous research on feedback has focused on analyzing different types of feedback and their effect on the learners’ writings and few studies have examined the effectiveness of computer feedback. Therefore, the present study was conducted to 1) determine what aspects of students’ writings receive computer feedback, 2) examine the difference in the effect of computer-generatedfeedback (CBF) and Teacher-based feedback (TBF) on improving the students’ writing quality and 3) compare the differences in Depth of Processing (DOP) in processing computer and teacher feedback. The results indicated that content, style and organization of their essays received feedback from the teacher and the computer. Teacher feedback was more effective in terms of its impact on improving the quality of the writing of the students than computer-generatedfeedback and it resulted in deeper processing of lexical items, whereas computer-generated feedback invoked medium processing on grammar.

The Effect of Blended Online Automated Feedback and Teacher Feedback on EFL Learners’ Essay Writing Ability and Perception

Applying both teacher- and computer-generated feedback to foster EFL learners’ writing remains a less explored issue. The purpose of this study was two-fold. Firstly, it examined whether blending both online automated feedback (OAF) and teacher feedback had a significant effect on EFL learners’ writing ability or not. Secondly, it explored their perceptions on the use of the blended OAF and teacher feedback. To this end,this study employed a mixed-method design using essay writing, questionnaire, and interview to investigate the impact of blended OAF and teacher feedback on EFL learners’ writing ability in an essay writing course. Thirty intermediate EFL learners participated in the study; they were given the first writing pre-test, and then were assigned into experimental and control groups randomly. The experimental group was given access to the integrated OAF and teacher feedback on their two-paragraph problem-solution essay writing during the treatment phase while the control group received teacher-only feedback. The results of the independent samples t-test and RM ANOVA showed that the experimental group outperformed the control group in terms of the overall score for the two final essay writings, and also demonstrated a significant improvement in writing scores across four essays during the treatment. The qualitative data also revealed the participants’ positive perception towards the effectiveness of the blended OAF and teacher feedback on their writing. Overall, the findings of this study can contribute to a better understanding of the impact of the blended feedback in enhancing EFL learners’ writing ability.

Exploring the effect of computer-mediated teacher feedback on the writing achievement of Iranian EFL learners: Does motivation count?

The use of a computer as a means and/or a source of feedback provision has facilitated the process of teaching and learning writing. The integration of computers into writing classes enabled teachers to provide timely and reliable feedback. Taking into account these opportunities that computers bring to the classroom, the present study attempted to investigate the efficacy of using computer-mediated teacher feedback and computer-generated feedback on learners’ writing skill. In addition, learners’ motivational level was explored. To do so, 60 intermediate EFL learners were selected from two intact classes and were randomly assigned to treatment groups. The results of paired samples t-test and independent samples t-test revealed a significant improvement in writing ability of the two groups from pretest to posttest. Only computer-mediated feedback significantly improved learners’ writing ability from posttest to delayed posttest. In addition, no significant difference was observed between the posttest scores of the two groups, while there was a significant difference between the delayed posttest score of the two groups. The result of interview indicated learners’ motivation in using computer-mediated feedback while there was disagreement in the other group concerning the motivation to use this approach. The study concluded that computers are a good medium for feedback provision. Furthermore, learners do not appreciate the role of computers as the source; rather, they accept it as a supplement to teacher feedback.

Print-based self-help interventions for smoking cessation.

Print-based self-help interventions for smoking cessation.

Can brief alcohol interventions in general hospital inpatients improve mental and general health over 2 years? Results from a randomized controlled trial.

Little is known about the impact of brief alcohol interventions on mental and general health. The aim was to investigate whether brief interventions for general hospital inpatients with at-risk drinking can improve mental and general health over 2 years; and whether effects are dependent on how they are delivered: in-person or through computer-generated feedback letters (CO). Three-arm randomized controlled trial with 6-, 12-, 18-, and 24-month follow-ups. Data were collected on 13 general hospital wards from four medical departments (internal medicine, surgical medicine, trauma surgery, and ear-nose-throat) of one university hospital in northeastern rural Germany. A consecutive sample of 961 18- to 64-year-old general hospital inpatients with at-risk alcohol use was recruited through systematic screening. Inpatients with particularly severe alcohol problems were excluded. Participants were allocated to: in-person counseling (PE), CO, and assessment only (AO). PE and CO included three contacts: on the ward, 1, and 3 months later. Mental and general health were assessed using the five-item mental health inventory (0-100) and a one-item general health measure (0, poor - 4, excellent). Latent growth models including all participants revealed: after 24 months and in contrast to AO, mental and general health were improved in PE (change in mean difference, ΔMmental = 5.13, p = 0.002, Cohen's d = 0.51; ΔMgeneral = 0.20, p = 0.005, d = 0.71) and CO (ΔMmental = 6.98, p &lt; 0.001, d = 0.69; ΔMgeneral = 0.24, p = 0.001, d = 0.86). PE and CO did not differ significantly. Beyond drinking reduction, PE and CO can improve general hospital inpatients' self-reported mental and general health over 2 years.

Word processors as monarchs: Computer-generated feedback can exercise power over and influence EAL learners' identity representations

While the efficacy of computer-generated feedback in affecting learners' scores, errors, and writing skills has already been established, the impact of such feedback on learners' identity representations remains unexplored. The current paper explores the ways in which computer-generated feedback from Microsoft Word Office™ (MWO) and Grammarly influences English as an Additional Language (EAL) learners' identity representations while writing on a screen. To this end three participants were involved in a case study whereby they were asked to complete five writing tasks. Power relations, intertextuality and multimodality levels, and attitudes—in their technical sense—as relevant concepts were employed to explore potential factors that influence EAL writers’ identity representations. The results obtained from think-aloud sessions and interviews show that while the participants relied on computer-generated feedback to spell correctly and to make well-formed sentences, they experienced pressures, control and power from automatic feedback, which subsequently influenced their identity representations.

In-person alcohol counseling versus computer-generated feedback: Results from a randomized controlled trial.

To investigate the comparative 2-year efficacy of brief alcohol interventions delivered in-person versus through computer-generated feedback letters among general hospital inpatients with at-risk alcohol use. In 2011-2012, all general hospital inpatients aged 18-64 years on 13 wards at 4 medical departments in 1 general hospital were systematically screened for at-risk alcohol use. Nine-hundred sixty-one inpatients who screened positive for at-risk alcohol use and negative for more severe alcohol problems were randomized by timeframe to in-person counseling (PE), computer-generated feedback letters (CO) and assessment only (AO). Both interventions included 3 contacts: on the ward and 1 and 3 months later and were based on the transtheoretical model (TTM). After 6, 12, 18, and 24 months computer-assisted telephone interviews assessed self-report outcomes: gram alcohol per week (primary), at-risk alcohol use and highest blood alcohol concentration (both secondary). Latent growth models were used. After 24 months, CO resulted in a greater reduction of self-reported gram alcohol per week than AO (p = .027); PE did not differ significantly from AO (p = .503) and CO (p = .088); and group differences concerning secondary outcomes were not statistically significant (ps > 0.07). After 6 months, the odds of at-risk alcohol use were reduced by half in PE versus AO (odds ratio = 0.50, 95% confidence interval: 0.25-0.98). No significant group differences between in-person counseling and computer-generated feedback letters were found. TTM-based computer-generated feedback letters reduced self-reported gram alcohol over 2 years and can be considered a long-term effective alternative in medical settings when addressing at-risk alcohol use. (PsycINFO Database Record

How alcohol use problem severity affects the outcome of brief intervention delivered in-person versus through computer-generated feedback letters

ObjectiveThe aim was to test if people with different alcohol use problem severity benefitted differentially from brief alcohol interventions delivered in-person versus through computer-generated feedback letters. MethodsNine hundred sixty-one 18–64year old general hospital inpatients with at-risk alcohol use (mean age=40.9years [standard deviation=14.1], 75% men) were randomized to a) in-person counseling, b) computer-generated individualized feedback letters, or c) assessment only. Both interventions were delivered on the ward and 1 and 3 months later. Outcome was the change in the self-reported alcohol use per day at 6-, 12-, 18-, and 24-month follow-up. The Alcohol Use Disorder Identification Test (AUDIT) score was tested as a moderator of the effect of in-person counseling and computer-generated feedback letters, with higher AUDIT scores indicating more severe alcohol problems. ResultsCompared to assessment only, computer-generated feedback letters more strongly reduced alcohol use over 24 months among persons with AUDIT scores of about 8 and lower (ps <0.05). In-person counseling tended to be superior over assessment only among persons scoring high on the AUDIT, but differences were not statistically significant. Six-, 12-, and 18-month differences between in-person counseling and computer-generated feedback letters were significant (ps<0.05) for persons with AUDIT scores below 7.1–7.7. The differences between both interventions attenuated at 24-month follow-up. ConclusionsComputer-based intervention delivery may be superior over in-person delivery for people with low levels of alcohol use problem severity, whereas those with more severe alcohol problems may require more intensive care.

Reconsidering Teacher, Peer, and Computer‐Generated Feedback

Reconsidering Teacher, Peer, and Computer‐Generated Feedback

Alcohol is an Unsafe Product

Alcohol is a hazardous product. Its consumption depends on the society's drinking norms. These are set by advertising industries on the one hand, and forces of public health and medical services on the other hand. Germany is one of the countries with high alcohol consumption. According to epidemiological evidence, alcohol-attributable diseases include more than 200 diagnostic entities. Medical practitioners may help to reduce alcohol consumption. Data revealed that brief intervention, including computer-generated feedback in medical practice, is efficient in reducing consumption. It relieves the physician of counseling responsibilities.

Mitigation of Cognitive Bias with a Serious Game

One of the benefits of using digital games for education is that games can provide feedback for learners to assess their situation and correct their mistakes. We conducted two studies to examine the effectiveness of different feedback design (timing, duration, repeats, and feedback source) in a serious game designed to teach learners about cognitive biases. We also compared the digital game-based learning condition to a professional training video. Overall, the digital game was significantly more effective than the video condition. Longer durations and repeats improve the effects on bias-mitigation. Surprisingly, there was no significant difference between just-in-time feedback and delayed feedback, and computer-generated feedback was more effective than feedback from other players.

Who benefits from computer-based brief alcohol intervention? Day-to-day drinking patterns as a moderator of intervention efficacy

Objectiveto test if people with different day-to-day drinking patterns benefitted differently from two brief alcohol interventions (BAIs). MethodsA total of 1243 job-seekers with at-risk alcohol use aged 18–64 years (64% men) were randomized to (a) intervention tailored to the motivational stage (ST), (b) non-stage tailored intervention (NST), or (c) assessment only (AO). ST and NST contained individualized computer-generated feedback letters. Follow-ups were conducted at months 3, 6, and 15. Using growth mixture models, day-to-day drinking patterns were identified based on the number of drinks consumed on each day in the week prior to baseline assessment. To test drinking pattern-specific intervention effects, zero-inflated growth models were used. Outcomes were (1) the 15-month change in the likelihood of any alcohol use and (2) the 15-month change in the total number of drinks per week when alcohol was consumed. ResultsFour day-to-day drinking patterns were found: daily medium use (2–4 drinks/day; 47%), daily low use (1–2 drinks/day; 29%), weekend only use (18%), and no use (6%). Only persons with daily low use benefitted from intervention, with higher odds of being abstinent after 15 months in the ST group compared to AO (odds ratio, OR=1.67, p=0.001) and NST group (OR=1.43, p=0.035). ST worked better among persons with daily low use compared to daily medium use (OR=1.91, p=0.001). ConclusionsAmong at-risk drinking persons with regular low-quantity alcohol use, stage tailored BAIs may be superior over no BAI and non-stage tailored BAIs.

The Contribution of Reasoning to the Utilization of Feedback from Software When Solving Mathematical Problems

This study investigates how students’ reasoning contributes to their utilization of computer-generated feedback. Sixteen 16-year-old students solved a linear function task designed to present a challenge to them using dynamic software, GeoGebra, for assistance. The data were analysed with respect both to character of reasoning and to the use of feedback generated through activities in GeoGebra. The results showed that students who successfully solved the task were engaged in creative reasoning and used feedback extensively.