Incorrect Mental Models Research Articles

Mental Models and Learning: The Case of Base-Rate Neglect

We experimentally document persistence of suboptimal behavior despite ample opportunities to learn from feedback in a canonical updating problem where people suffer from base-rate neglect. Our results provide insights on the mechanisms hindering learning from feedback. Importantly, our results suggest mistakes are more likely to be persistent when they are driven by incorrect mental models that miss or misrepresent important aspects of the environment. Such models induce confidence in initial answers, limiting engagement with and learning from feedback. We substantiate these insights in an alternative scenario where individuals involved in a voting problem overlook the importance of being pivotal. (JEL D83, D91)

THE IMPACT OF COMBINING FOLLOW-UP QUESTIONS AND WORKED EXAMPLES IN PROGRAM VISUALIZATION TOOL ON IMPROVING STUDENTS’ HELD MENTAL MODELS OF POINTERS’ VALUE AND ADDRESS ASSIGNMENT

Previous studies have shown that the lack of a useful mental model of pointers is one of the reasons why many novice programmers fail the data structures course. This study had two main objectives: to analyze the status of mental models of pointers (focusing on value and address assignment); and to evaluate the impact of combining worked-examples and follow-up questions in CeliotM program visualization (PV) tool in the learning of pointers. The subjects of the study were sixty-two second-year undergraduate students taking a course on data structures (PMT 221) at the College of Natural and Mathematical Sciences (CNMS) of the University of Dodoma. Data were collected using pretest and posttest questionnaires. The collected data were analyzed using descriptive statistics. The results showed that 56.5% of the students had incorrect mental models of pointers. The results also showed that using the proposed strategy improved the students’ mental models of pointers from 56.5% to 87.1%. These results contribute to our understanding of the most common misconceptions that novice students may have when learning pointers. The findings of this study confirm previous studies that when the new innovative teaching strategies are used in combination with PV tools in teaching and learning programming can help improve students’ programming comprehension.

When Mental Models Grow (C)old: A Cognitive Perspective on Home Heating Automation

Have you ever tried to speed up a revolving door by pushing the panel in front of you, only to trigger its security mechanism and make it stop? If so, it is a simple example of a situation in which your mental representation of a system (door) and its actual functioning did not align. In other words, you applied an incorrect mental model. Understanding mental models can be important in a variety of contexts, including smart homes, where they impact next-generation heating control.

What You See Is All There Is*

Abstract News reports and communication are inherently constrained by space, time, and attention. As a result, news sources often condition the decision of whether to share a piece of information on the similarity between the signal and the prior belief of the audience, which generates a sample selection problem. This article experimentally studies how people form beliefs in these contexts, in particular the mechanisms behind errors in statistical reasoning. I document that a substantial fraction of experimental participants follows a simple “what you see is all there is” heuristic, according to which participants exclusively consider information that is right in front of them, and directly use the sample mean to estimate the population mean. A series of treatments aimed at identifying mechanisms suggests that for many participants, unobserved signals do not even come to mind. I provide causal evidence that the frequency of such incorrect mental models is a function of the computational complexity of the decision problem. These results point to the context dependence of what comes to mind and the resulting errors in belief updating.

What You See is All There Is

News reports and communication are inherently constrained by space, time, and attention. As a result, news sources often condition the decision of whether to share a piece of information on the similarity between the signal and the prior belief of the audience, which generates a sample selection problem. This paper experimentally studies how people form beliefs in these contexts, in particular the mechanisms behind errors in statistical reasoning. I document that a substantial fraction of experimental participants follows a simple “what you see is all there is” heuristic, according to which they exclusively take into account information that is right in front of them, and directly use the sample mean to estimate the population mean. A series of treatments aimed at identifying mechanisms suggests that for many participants unobserved signals do not even come to mind. I provide causal evidence that the frequency of such incorrect mental models is a function of the computational complexity of the decision problem. These results point to the context-dependence of what comes to mind and the resulting errors in belief updating.

The evolution of mental models in relation to initial information while driving automated

ObjectiveMental models guide drivers’ expectations about the functioning of a conditionally automated vehicle. We induced different mental models using preliminary system descriptions to explore mental models with an objective, online measurement during conditionally automated driving. BackgroundHuman-machine interaction, based mainly on mental models, has been examined mostly by employing subjective measurements. However, an objective measurement method could improve the comparability of studies and provide a broader understanding of mental models in automated driving, thus leading to road safety in times of mixed traffic. Methods and resultsIn two experiments (total N = 148), we manipulated the participants’ mental models by providing correct and incorrect system descriptions. In Experiment 1, contrary to our expectations, the results showed faster reaction times in the condition with an incorrect mental model. In Experiment 2, we replicated this finding and showed that this effect can be traced back to the very first experience of the mismatch between the mental model and actual system behavior. ConclusionOverall, our results showed an impact of the preliminary system description on mental models. Moreover, the importance of a complete and correct manipulation of materials in conditionally automated driving research is emphasized. ApplicationPotential applications include the online assessment of mental models during automated driving (e.g. dead man’s switch).

Teaching distributed memory programming from mental models

Distributed memory programming, typically through the MPI library, is the de facto standard for programming large scale parallelism, with up to millions of individual processes. Its dominant paradigm of Single Program Multiple Data (SPMD) programming is different from threaded and multicore parallelism, to an extent that students have a hard time switching models. In contrast to threaded programming, which allows for a view of the execution with central control and a central repository of data, SPMD programming has a symmetric model where all processes are active all the time, with none privileged, and where data is distributed.This model is counterintuitive to the novice parallel programmer, so care needs to be taken how to instill the proper ‘mental model’. Adoption of an incorrect mental model leads to broken or inefficient code.We identify problems with the currently common way of teaching MPI, and propose a structuring of MPI courses that is geared to explicit reinforcing the symmetric model. Additionally, we advocate starting from realistic scenarios, rather than writing artificial code just to exercise newly-learned routines.

The role of system description for conditionally automated vehicles

ObjectiveWe studied how system descriptions of conditionally driving vehicles (SAE International, 2014) influenced drivers' knowledge/mental model, trust, and acceptance. BackgroundThe increasing proliferation of assisted and conditionally automated vehicles urge a proper understanding of knowledge about, acceptance, and trust in the system. Up to now there is a lack of studies for automated systems and system knowledge which will be served in this study. MethodsWe provided N = 120 participants with correct, incomplete, and incorrect descriptions about the conditionally automated vehicle. Thereafter, they watched five video clips with different situations where the driver has to resume control of the driving task or do nothing. Immediately before such a situation, the videos were frozen and knowledge questions were asked according to the SAGAT method (Endsley, 1995). Further, trust and acceptance before and after watching videos were measured. ResultsThe knowledge questions differed in groups for level 2 (comprehension) and level 3 (projection) of situation awareness (SA). The difference is significant for the groups in level 2 situation awareness (comprehension) with correct vs. incorrect and incomplete vs. incorrect descriptions. The overall trust in their answer behavior was the highest for the group with the correct description followed by the group with the incomplete description. Furthermore, trust and acceptance did not differ between groups and even not between measurement time (before vs. after watching videos). ConclusionAn incorrect preliminary system description leads to an incorrect mental model which in turn leads to incorrect comprehension and projection of traffic situations. This effect can be minimized over time, but is relevant to the safety of the driver. ApplicationAccurate system descriptions might reduce the number and the gravity of most accidents with conditionally automated systems in driving. Moreover, repeated exposure will reduce the possibility of accidents.

Systems thinking in 21st century: a call to health promoters

<no abstract available>

One output function: a misconception of students studying digital systems – a case study

Background:Learning processes are usually characterized by students’ misunderstandings and misconceptions. Engineering educators intend to help their students overcome their misconceptions and achieve correct understanding of the concept. This paper describes a misconception in digital systems held by many students who believe that combinational logic circuits should have only one output.Purpose:The current study aims to investigate the roots of the misconception about one-output function and the pedagogical methods that can help students overcome the misconception.Sample:Three hundred and eighty-one students in the Departments of Electrical and Electronics and Mechanical Engineering at an academic engineering college, who learned the same topics of a digital combinational system, participated in the research.Design and method:In the initial research stage, students were taught according to traditional method – first to design a one-output combinational logic system, and then to implement a system with a number of output functions. In the main stage, an experimental group was taught using a new method whereby they were shown how to implement a system with several output functions, prior to learning about one-output systems. A control group was taught using the traditional method. In the replication stage (the third stage), an experimental group was taught using the new method. A mixed research methodology was used to examine the results of the new learning method.Results:Quantitative research showed that the new teaching approach resulted in a statistically significant decrease in student errors, and qualitative research revealed students’ erroneous thinking patterns.Conclusions:It can be assumed that the traditional teaching method generates an incorrect mental model of the one-output function among students. The new pedagogical approach prevented the creation of an erroneous mental model and helped students develop the correct conceptual understanding.

Educating Engineering Undergraduates: Effects of Scaffolding in a Problem-Based Learning Environment

This paper presents a study in the methods of problem-based learning (PBL) in the engineering classroom, where students work in small groups to explore specific problems under the guidance of an instructor. PBL has proven to be highly-effective in engineering education, but there is still room to improve. Studies by Schmidt et al. (2007) suggest that the use of “scaffolds,” structures that support the conceptual learning process early-on but are gradually removed later, can greatly help when students first engage in PBL. A total of 19 students participated in the study. A detailed lab handout was designed to act as a hard scaffold for the experimental group while the control group used the current materials. After preparing a full lab report, students were given a post-lab examination. The results of this examination indicate that the scaffold was effective, but they also hint that students enter the lab with incorrect mental models which hinder the learning process. Further study is needed to validate this hypothesis.

Communication and knowledge sharing in human–robot interaction and learning from demonstration

Inexpensive personal robots will soon become available to a large portion of the population. Currently, most consumer robots are relatively simple single-purpose machines or toys. In order to be cost effective and thus widely accepted, robots will need to be able to accomplish a wide range of tasks in diverse conditions. Learning these tasks from demonstrations offers a convenient mechanism to customize and train a robot by transferring task related knowledge from a user to a robot. This avoids the time-consuming and complex process of manual programming. The way in which the user interacts with a robot during a demonstration plays a vital role in terms of how effectively and accurately the user is able to provide a demonstration. Teaching through demonstrations is a social activity, one that requires bidirectional communication between a teacher and a student. The work described in this paper studies how the user's visual observation of the robot and the robot's auditory cues affect the user's ability to teach the robot in a social setting. Results show that auditory cues provide important knowledge about the robot's internal state, while visual observation of a robot can hinder an instructor due to incorrect mental models of the robot and distractions from the robot's movements.

The Mismatch between Students’ Mental Models of Acids/Bases and their Sources and their Teacher’s Anticipations thereof

The aim of this study is to compare the characteristics and sources of students’ mental models of acids and bases with a teacher’s anticipations and, based on this comparison, to explore some possible explanations why motivated students might fail to learn from a subject‐knowledgeable chemistry teacher. The study involves a chemistry teacher and her 38 ninth graders and focuses on the mental models of three high achievers and three low achievers who were interviewed in depth. Four students’ mental models of acid and base are identified. The mental models and sources of students’ conceptions of acids and bases that influenced the high achievers are compared to those of the low achievers. We find that the teacher in the study made accurate anticipations of her students’ mental models in the case of the high achievers but inaccurate anticipations of the low‐achievers’ mental models and the diverse sources influencing their mental models. In addition, the teacher incorrectly attributed the poor achievement of the low‐achieving students to their intuition and underestimated the effects of her teaching on the achievement of these students. As a result, the teacher’s instruction reinforced the low‐achievers’ incorrect mental models. Finally, the different approaches for teaching students with different achievements are emphasized according to the empirical data in this study.

Top-down and bottom-up influences on learning from animations

To evaluate how top-down and bottom-up processes contribute to learning from animated displays, we conducted four experiments that varied either in the design of animations or the prior knowledge of the learners. Experiments 1–3 examined whether adding interactivity and signaling to an animation benefits learners in developing a mental model of a mechanical system. Although learners utilized interactive controls and signaling devices, their comprehension of the system was no better than that of learners who saw animations without these design features. Furthermore, the majority of participants developed a mental model of the system that was incorrect and inconsistent with information displayed in the animation. Experiment 4 tested effects of domain knowledge and found, surprisingly, that even some learners with high domain knowledge initially constructed the incorrect mental model. After multiple exposures to the materials, the high knowledge learners revised their mental models to the correct one, while the low-knowledge learners maintained their erroneous models. These results suggest that learning from animations involves a complex interplay between top-down and bottom-up processes and that more emphasis should be placed on how prior knowledge is applied to interpreting animations.

TIPS (trigger an IIF paradigm shift)

Challenge New Mexico Corporate Services (NMCS) recordable injuries have been averaging 2–4 per year for the last 5 years with no statistical improvement. However, we believe all NM employees must go home incident and injury free every day and just as healthy as they came to work. In addition, we have received feedback from several sources, that indicates some employees are reluctant to report injuries. These indicators showed us that continuing our current strategies, making incremental improvement and changes, would not give us the improvement desired. We needed a paradigm shift to get everyone completely engaged in the IIF (Incident & Injury Free) culture, in order to achieve true IIF results. Methods/Strategies We formed a small (3-person) taskforce consisting of safety representatives from EHS, Site Services and CS Operations. We reviewed 5 years worth of data to determine what was injuring our people. We also decided to review all injuries, not simply those classified as recordable by OSHA standards. First we identified the types of injury information needed to get a true picture of our safety issues. We analyzed IRB (Incident Review Board) data showing us the following factors and whether any of them contributed to the injury: – Date – Incident Description – Severity – Root Cause – Type of Injury – Season – Work Group/Shift – Area – Improper evaluation of hazard – Inadequate work procedures – Incorrect Mental Model – Inadequate PPE Requirement – Failure to Follow or Unaware of PPE Requirement – Shortcut or Schedule Pressure – Last or First Day of Shift or Adjacent to Holiday – OT – Aggravate Existing Condition – Inadequate Training or Passdown – Experience in Task – Corrective Action Taken – Overall Quality of Response. Once this information was collected for all injuries in an Excel file, we graphed it several ways to help reveal trends: - Shift 7 had double the injuries of shift 5 - Night shift injuries were relatively high but lower than Shift 7 - Shift 5 had no severe (recordable) injuries - Biggest total injury type was ergonomic - Biggest recordable injury type (58%) was cuts/lacerations - Chemical exposure and inhalations resulted in no recordables - Biggest root cause category was behavioral - Biggest behavioral root cause was “Failure to Evaluate Hazards” - Biggest administrative controls root cause was “Inadequate Procedures” - Biggest engineering controls root cause was “Unrecognized Workplace Hazard” - Majority of injuries occurred in Spring/Summer which accounted for 86% of recordables - Response was inadequate or poor for 25% of injuries - Number of injuries by workgroup were about equal - Work area injuries were mostly in the CUB and Subfab but CUB injuries favored cuts/lacerations while Subfab favored ergonomic injuries These learnings and gaps were used to develop our New Mexico Site Safety Action Plan for 2005. It has also been used by individual managers and supervisors in their day-to-day business meetings, activities, and communications, for a more focused IIF message and effort. Results ▪ Conclusions/Recommendations Challenges are different amongst shifts. Shift culture may be stronger than workgroup culture. Work areas pose different challenges that all should be aware of and prepared for. We should prepare ourselves for challenges posed by the Spring and Summer seasons as well. We should be aware of and prepare for the different challenges and cultures characteristic of each shift. Supervisors need to spend more time in the field with their people to help them address safety issues and help make IIF a more consistent component of daily workgroup culture. Each site should analyze their data and look for their unique safety challenges. This will allow customized, focused action plans that meet the unique needs of each site.

The Development of Cognitive Skills To Support Inquiry Learning

Establishing the value of inquiry learning as an educational method, it is argued, rests on thorough, detailed knowledge of the cognitive skills it is intended to promote. Mental models, as representations of the reality being investigated in inquiry learning, stand to influence strategies applied to the task. In the research described here, the hypothesis is investigated that students at the middle school level, and sometimes well beyond, may have an incorrect mental model of multivariable causality (one in which effects of individual features on an outcome are neither consistent nor additive) that impedes the causal analysis involved in most forms of inquiry learning. An extended intervention with 6th to 8th graders was targeted to promote (a) at the metalevel, a correct mental model based on additive effects of individual features (indicated by identification of effects of individual features as the task objective); (b) also at the metalevel, metastrategic understanding of the need to control the influences of other features; and (c) at the performance level, consistent use of the controlled comparison strategy. Both metalevel advancements were observed, in addition to transfer to a new task at the performance level, among many (though not all) students. Findings support the claim that a developmental hierarchy of skills and understanding underlies, and should be identified as an objective of, inquiry learning.

Making sense of how students make sense of mechanical waves

We have investigated student difficulties with mechanical waves and used the idea of mental models to organize our findings. Mental models describe the analogies and guidelines used to come to understand a topic. We find that individual students use both correct and incorrect mental models to make sense of a single physical situation.

Mental Models of Recursion

This article presents the results of a two-year study of children's difficulties in learning the concept of recursion in Logo. A number of incorrect mental models of recursion identified in the study are described. These are classified into main groups: mental models associated with the interpretation of recursive procedures and those associated with the construction of recursive procedures. Misconceptions underlying the erroneous mental models are presented. Possible reasons for adopting these misconceptions are put forward.