how to understand research articles

.docx

School

Vanderbilt University *

*We aren’t endorsed by this school

Course

2200

Subject

Chemistry

Date

Dec 6, 2023

Type

docx

Pages

Uploaded by DrIbis2028

Understanding Primary Research Articles As you are probably already aware, primary research articles are more difficult to read and understand than textbooks. Below are a set of guidelines for how to read such articles and what kinds of notes to take while you’re reading. Our discussion of the research articles in class will start with these questions. Please come to class prepared to answer them. 1. Identify the central research question. What is/are the researcher(s) trying to understand by conducting the experiment(s) reported in this journal article? The researchers were trying to determine if medical experts who were exposed to medical images were able to form an impression of normality and abnormality and the nature of the associated global signal. 2. What is the design of the study? That is, what independent variable(s) was/were manipulated, and what are the different levels of each independent variable? Were the different levels of the/each IV presented between subjects (e.g., different subjects heard words spoken in a New York accent versus a Deep South accent) or within subjects (e.g., everybody heard one list of words spoken in a New York accent and another list of words spoken in a Deep South accent)? This is a within subjects – every subject was exposed to images in a mixture of two durations. The independent variable was how long the subjects were exposed to the pictures. There were different levels of this, some subjects would see more images for shorter exposure durations and vice versa. 3. Who were the subjects? What subject population was recruited for this research? Why this subject population rather than some other subject population? There were 55 radiologists, 38 cytologists, and 60 nonexpert control observers. The radiologists and cytologist have years of medical expertise with looking at medical images – the researchers wanted to see how background knowledge and expertise would affect their abiliy to remember and encode information from the pictures. 4. What was the subjects’ task? What procedure did the subjects follow? Subjects were tasked with viewing medical images (mammograms or cytology images) for very brief durations ranging from 250 to 2,000 ms. Their primary task involved rating the abnormalities present in the images and, in some experiments, localizing abnormalities by clicking on the display screen. The task required participants to form a global impression of normality or abnormality within the short exposure time. Additionally, the researchers examined the subjects' ability to perform these tasks across different exposure durations and expertise levels. 5. What is the dependent variable (DV)? In other words, how is subjects’ performance on the experimental task assessed? Note that a study may have more than one DV. The dependent variables in this study include d' values derived from confidence ratings, representing sensitivity in discriminating abnormalities; the area under curve (AUC) of the Receiver Operating Characteristic (ROC) curve calculated from these ratings; and the percentage of observers' clicks inside predetermined regions of interest, assessing localization accuracy for trials with correct abnormality ratings. These measures collectively gauge subjects' performance in rapidly detecting, categorizing, and localizing abnormalities in medical images within brief exposure

durations. The study employs a comprehensive set of dependent variables to capture various aspects of the subjects' performance on the experimental tasks. 6. What is the hypothesis or prediction? That is, which condition(s) (i.e., levels of the independent variable) are expected to lead to better performance? Why? The researchers' hypotheses, although not explicitly stated, may involve expectations that expert radiologists and cytologists possess the ability to rapidly form a global impression of normality or abnormality in medical images, leveraging an implicit global signal based on image statistics. Additionally, there might be a hypothesis that this global signal, while contributing to rapid gist extraction, is insufficient for precise localization of abnormalities. The researchers could also anticipate that the gist signal, if identified by a computer, could be utilized as a novel form of computer-aided detection, serving as a warning of a statistical signal associated with an elevated likelihood of a positive finding. These interpretations are speculative and would require direct confirmation from the researchers' explicit hypotheses in the study. 7. What were the results of the study? In which conditions did subjects do better, worse? How did the independent variables affect performance on the dependent variable(s)? The study revealed that both expert groups, radiologists and cytologists, performed above chance in detecting subtle abnormalities across different exposure durations, with sensitivities (d' values) around 1.2. Nonexpert control groups did not surpass chance performance, especially at shorter durations. Longer exposures significantly improved expert performance, as demonstrated by radiologists and cytologists. Despite this rapid gist extraction ability, both expert groups showed poor performance in localizing abnormalities, with no improvement as confidence ratings increased. 8. Were the hypotheses supported? Explain the link between the results and your conclusions in this regard. The study's hypotheses were generally supported by the results. Both expert groups exhibited above-chance performance in rapidly detecting subtle abnormalities, consistent with the hypothesis that experts could form a global impression of normality or abnormality. The observed improvement in performance with longer exposure durations aligned with the expectation that extended exposure facilitates better abnormality detection. However, the poor performance in localizing abnormalities contradicted the hypothesis that the global signal alone would be sufficient for precise localization. This suggests that while the global signal contributes to rapid gist extraction, it does not effectively guide precise localization, revealing certain limitations in the hypotheses. 9. IDEAS FOR DISCUSSION: For example: What further questions do you have about this research? What did you find interesting about this research? How would you follow up on the results of this study with new research? How are the results of this research applicable in other domains? To enhance the study, incorporating eye tracking technology and neuroimaging techniques could elucidate the interplay between the global signal and eye movements, while exploring diverse image types and conducting longitudinal studies would provide a more comprehensive understanding of experts' rapid image processing abilities and their potential applications in clinical settings.

Your preview ends here

Eager to read complete document? Join bartleby learn and gain access to the full version