The Eagle Eyes: an Intervention Utilizing Visual Thinking Strategies to Enhance the Observation Skills of Medical Students

Abstract

Background

Visual Thinking Strategies (VTS) is an evidence-based pedagogical approach that uses art analysis and structured inquiry to enhance observation, critical thinking, and teamwork, especially in medical training for clinical skills development. This study aimed to compare the short-term and delayed follow-up effects of integrating Visual Thinking Strategies and Visual Thinking Activity (VTA) tasks based on the PRISM Model with Observation Exercises (OE) on medical students’ observation skills, including the number of observations, number of words used, and time spent describing observations.

Method

This pre- and post-test experimental study with a control group was conducted among first-year medical students at Gonabad University of Medical Sciences during the 2023–2024 academic year. Forty-four students participated in the intervention group, receiving VTS and VTA tasks, while 45 students formed the control group, engaging in OE alone. Observation skills were assessed using standardized art images (short-term) and real-world clinical exposure (delayed follow-up) through measures of total observations, number of words used, and time spent describing observations. Descriptive statistics, analysis of one-way ANOVA/ANCOVA, and independent t-tests were employed for data analysis.

Results

In the short-term evaluation, the intervention group demonstrated significantly higher performance in the total number of observations (p = 0.001, Adjusted Partial Eta2 = 0.12), number of words used to describe art images (p = 0.001, Adjusted Partial Eta2 = 0.21), and time spent analyzing images (p < 0.001, Adjusted Partial Eta2 = 0.17) compared to the control group. However, after one month in a clinical exposure, no significant differences were found between the groups in the total number of observations (p = 0.62) and number of words used (p = 0.64). Nevertheless, the intervention group spent significantly more time describing their clinical observations (p = 0.04, Effect Size = 0.44).

Conclusion

The findings highlight the significant role of VTS in enhancing medical students’ observation skills. While both interventions were equally effective in the delayed follow-up and real-world settings regarding the total number of observations and words used, the VTS and VTA approach led to a notable increase in the time spent on observation descriptions. This conclusion warrants further investigation in future studies.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12909-025-06642-9.

Introduction

In recent decades, the medical education research community has increasingly recognized the pivotal role of observation skills among the core competencies required of medical students [1]. Observation is a fundamental tenet of the medical discipline, serving as a crucial precursor to diagnosis, prognosis, and treatment [2]. Physicians’ ability to conduct meticulous patient examinations and guide the diagnostic process hinges upon their observational prowess [2]. However, while patient observation is an essential and learnable skill, it is rarely the subject of explicit instruction within medical curricula [2–4].

Even though the observational acumen that distinguishes the seasoned, astute clinician is typically acquired only through years of extensive clinical experience, most medical schools and residency programs devote little attention to the systematic development of this critical competency [3, 5–7]. Consequently, medical students are often criticized for lacking this essential skill set [3, 5–7].

Existing evidence suggests that teaching observation skills through traditional instructional approaches in medical education presents significant challenges [8, 9]. In contrast, art-based education offers a promising alternative, providing formal and explicit training in observation and description skills [6, 8]. Integrating art-based curricula has been shown to enhance student engagement and lead to improved medical performance [6, 8, 9]. Incorporating observation skill development into art-based educational frameworks may offer a viable solution to the existing educational dilemmas surrounding this crucial competency [10].

The integration of arts-based curricula into medical education has garnered significant attention in recent years. However, the extant literature lacks comprehensive information on the impact of these arts-based interventions on medical students’ behaviors, attitudes, and technical skills [11, 12]. To justify the inclusion of arts-based courses as an integral component of medical education, it is imperative that studies rigorously assess the effectiveness of such interventions [11, 12]. Recognizing the potential value of integrating the arts and humanities into medical training, major professional organizations have increased their focus on developing and implementing these novel educational approaches [13–15]. The Association of American Medical Colleges (AAMC) and the National Academy of Sciences, Engineering, and Medicine (NASEM) have actively supported the advancement of arts- and humanities-based courses, emphasizing the need to evaluate “learner outcomes beyond satisfaction with the course or program” [16, 17].

The integration of visual arts into medical curricula through Visual Thinking Strategies is increasingly recognized for its educational significance [8, 18]. Salvatore Mangione highlights VTS’s emphasis on detail-oriented analysis and deductive reasoning, akin to medical methodologies [7]. Devised by Abigail Housen and Philip Yenawine [8, 19, 20], VTS employs an evidence-based model where participants collaboratively scrutinize artworks. The facilitator prompts this analysis with three structured inquiries: “What’s going on in this image?”, “What do you see that makes you say that?”, and “What more can you find?” [8, 12, 19, 21–24]. These questions are designed to elicit detailed observations and foster a group dialogue, thereby enhancing various competencies [19, 24] and promoting a collective interpretative process [8, 19, 20, 23, 25].

Considering the inherent complexities of medical education and learning, the integration of arts and humanities into the field presents significant challenges. To address this, the Association of American Medical Colleges (AAMC) has introduced the Prism model, a novel framework for integrating arts and humanities programs into medical education [16, 26–29]. This framework aligns with the six core competencies of the Accreditation Council for Graduate Medical Education (ACGME) [16, 28, 29]. The Prism model outlines four key functions of arts and humanities in medical education: Mastering Skills, Perspective Taking, Personal Insight, and Social Advocacy. On the one hand, improving observational and judgment skills fall under the Patient Care category, aiding patient evaluation skills, medical decision-making, and intra-operative technical skills [30, 31]. Medical educators can utilize the Mastering Skills lens within the Prism model to develop and strengthen these essential skills [16, 28, 29]. Studies suggest that VTS, an evidence-based approach to engaging students with art, have the potential to support the growth of observational, communicative, interpersonal, and diagnostic skills [9, 16, 32]. Additionally, VTS may contribute to perspective-taking [9, 28] and personal insight [33] as students reflect on their own views, aligning with the Prism model’s corresponding functions [11, 34, 35]. However, the extent and manner in which arts and humanities programs like VTS achieve these educational outcomes remain unclear [34].

Few studies demonstrate that VTS can enhance observational skills through the duration students spend analyzing an image, the number of observations made, and the length of responses [18, 36, 37]. A study by Klugman et al. demonstrated that exposure to VTS increased the time spent observing clinical images, word count, and number of observations made by the participants [36]. However, this study did not analyze the quality of the observations and lacked a control group. Similarly, Naghshineh et al. conducted a controlled study with a significant number of VTS sessions (eight in total), each with a distinct educational component separate from the VTS training [37]. Although this study found positive outcomes, it did not analyze the duration of observations, as a predetermined time of eight minutes was allocated for the students to complete their clinical and art images descriptions. More recently, Agarwal et al. employed a pre-test post-test study design to evaluate the impact of two VTS workshop sessions on the observational skills of first-year medical students [18]. This study utilized a control group and assessed the effect of VTS alone, finding that VTS training increased the total word count used to describe clinical images, the time spent analyzing the images, and the number of clinically relevant observations made. Unlike the previous studies, Agarwal et al. included a broader range of clinical images, such as electrocardiograms and radiology images, in addition to physical findings. The existing research has failed to address the long-term follow-up of program outcomes, considering the continuity of impact.

The present study employed a pre-test and post-test design to evaluate the effect of a four-session VTS course on the observational skills of first-year medical students. Building upon the existing research by Naghshineh, Klugman, and Agarwal, this study aimed compare the short-term and delayed follow-up effects of integrating VTS and Visual Thinking Activity tasks based on the PRISM Model with Observation Exercises on medical students’ observation skills, including the number of observations, number of words used, and time spent describing observations.

Method

Study design

The study was conducted at the Gonabad University of Medical Sciences, Iran, during the academic year 2023–2024, with the approval of the institutional review board (ethics code: IR.TUMS.MEDICINE.REC.1402.094) and Iranian Registry of Clinical Trials code (IRCT20240716062444N1). All first-year medical students enrolled at the university were invited to participate in the study. After providing informed consent, the participants completed a baseline survey to collect demographic and background information. They then underwent a pre-test assessment, which involved analyzing and describing a set of art images. The total number of observations made, the number of words used, and the time spent on the task were recorded as the primary outcome measures. The participants were randomly assigned to either the intervention or control group. The intervention group received a four-session VTS course, which integrated the VTS methodology with Visual Thinking Activities (VTAs) based on the PRISM model. An initial draft of the Intervention design utilizing the PRISM model was composed and shared with the educational design team, and was approved for implementation (Appendix A). The control group, on the other hand, participated in a series of OE. Approximately one month later, an early clinical exposure session was conducted for both groups to determine the practical application of the training provided in the real setting. The clinical exposure for the control and intervention groups was conducted on separate days, with the control group on the first day and the intervention group on the second. Students rotated in groups to observe the surgical ward environment and patients.

Setting

The study was conducted at the Gonabad University of Medical Sciences (GMU), where the undergraduate medical education program consists of a seven-year curriculum. The program is divided into four main stages: basic sciences (two years), pathophysiology (one year), clerkship (two and a half years), and internship (one and a half years). The general medical curriculum in the clerkship and internship phases follows an organ-based approach. It is important to note that the current medical curriculum at GMU does not provide formal instruction on the skills of “observing and describing.” Instead, the emphasis is on teaching students to recognize normal and abnormal findings, particularly the cardinal signs and symptoms of disease. The curriculum focuses on the memorization of clinical signs, with less emphasis on the actual skill of careful observation.

Sampling

To select the control and intervention groups, the study utilized the existing student grouping upon entry into the medical school. At GMU, first-year medical students are divided into two groups based on their intake: September intake and January intake. For this study, the September intake students were randomly assigned to the intervention group, while the January intake students were randomly assigned to the control group (Fig. 1). This approach was chosen to minimize the potential for cross-contamination of the intervention within the study sample. Therefore, the study is controlled but not randomized. All participants were required to complete an informed consent form and a baseline survey through ePoll, an online survey management system. The baseline survey assessed educational backgrounds and clinical experiences. Neither group had formal education in the humanities or arts, nor any prior clinical experience that could impact the study’s outcomes.

Fig. 1.

The flow of the experiment

Intervention group

The intervention group participated in a four-week VTS training course from April 2 to May 21, 2023, which consisted of four weekly sessions, each lasting two and a half hours. The course was facilitated by a single facilitator who was familiar with the VTS methodology and had a strong interest in visual arts. The research team carefully selected nine artworks based on predefined criteria. These criteria were designed to ensure that the selected images were suitable for the VTS approach and would effectively engage the participants. The art pieces included in the selection were those with a possible connection to healthcare and portrayed images of the ill, distressed, or social determinants of health [38] Three artworks were discussed in each session.

The team’s criteria for selecting the images were as follows [18, 19, 38]:

1. Interpretability without specialized knowledge: The images were chosen to be interpreted without requiring specialized art knowledge, as long as the participants engaged in careful observation and thoughtful discussion.

2. Relevance to healthcare and medical sciences: Some of the selected images were directly related to healthcare and medical sciences, to ensure relevance to the participants’ field of study.

3. Presence of ambiguity: The images contained elements of ambiguity, which is a crucial aspect of the VTS approach as it encourages a diversity of perspectives and collaborative exploration.

4. Focus on narrative art without abstract concepts: The content of the selected images was narrative, without relying on abstract objects or concepts.

5. Potential for multiple interpretations and collaborative exploration: The images were chosen to facilitate multiple interpretations and encourage collaborative group discussions.

Each training session in the four-week VTS course followed a similar structure. The session began with a 90-minute guided group discussion based on the principles of VTS. During this discussion, students were encouraged to pay close attention to their communication and interaction patterns, engaging in a process of metacognition. After the VTS discussion, the students were invited to reflect on the process and evaluate their performance. They were prompted to consider questions such as “What did we do?“, “How did we work together as a group?“, “What did we notice?“, and “What did we particularly enjoy and find interesting about this process?” The discussion also emphasized the value of listening to different perspectives, the importance of active listening, and the benefits of refraining from hasty judgments. This approach allowed the students to view art as an ongoing, open-ended process [39]. Following a short break, the session continued with one of the Visual Thinking Activities (VTAs). These activities focused on applying observation skills through various exercises and facilitated the students’ understanding of the role of the arts in comprehending patient experiences [40].

Over the four training sessions, a different VTS-based art activity was incorporated into the final 60 min of each session. These activities included a Personal Responses Tour, Group Poems, Describe and Draw and Mask Making. Throughout the intervention, the VTS-based discussions and visual thinking activities were recorded with the participants’ consent. It is important to note that all three training sessions utilized art or art-based images with medical connections to practice observation based on VTS. In the fourth session, the students applied the VTS process to a live human model. The details of the intervention are provided in Appendix B.

The first training session began with a 90-minute VTS-guided group discussion focused on three pieces of art or art-based images with medical connections. Following the group reflection and a short break, the “Personal Responses Tour” activity was introduced. A specific artwork, “Watson and the Shark” by John Singleton Copley (1778), was presented, and participants were asked to discuss how the piece relates to an aspect of their current professional life. They were also prompted to consider the challenges and benefits of new professional roles, and to write a line of dialogue from the perspective of a character depicted in the artwork. The students worked collaboratively in small groups to complete this task and then shared their responses with the larger group [23, 34, 39].

In the second session, a similar 90-minute VTS-guided group discussion was conducted, this time focusing on three different art pieces or art-based images with medical connections. After the group reflection and a short break, the “Group Poems” activity was introduced. Students were divided into small groups and tasked with collaborating to create a written piece based on a visual prompt. Each group analyzed a pre-selected artwork depicting at least one figure and wrote a phrase they believed the figure might say. The groups then worked together over the next 20 min to arrange the lines into a song or poem, which they subsequently performed for the class and engaged in a group discussion [23].

The third training session followed the same structure, with a 90-minute VTS-guided group discussion on three art pieces or art-based images with medical connections. After a brief break, the “Describe and Draw” (Back-to-Back) activity was introduced. In this exercise, pairs of students received a pre-selected artwork. One person, the Describer, looked at the artwork and described it to the other person, the Drawer, who could not see the original. The Drawer then drew what they imagined the artwork looked like based on the Describer’s description. Following the activity, the pairs reflected on the challenges of being a Describer and a Drawer, their ability to listen or describe during the task, and discussed ways to improve the accuracy and descriptiveness of the process [29, 41, 42].

In the fourth and final session, the initial 40 min were dedicated to applying VTS to a human model, simulating clinical conditions such as allergic skin rashes, bruises, wounds, sutures, and burns [40]. This was followed by a 20-minute VTS discussion on a clinical image of a patient with skin conditions, including port-wine stains (birthmarks), with the expertise of a dermatologist consulted. The session then transitioned to a group discussion and reflection on the experience of using VTS with a live human model and a clinical image, compared to the artwork presented in the previous sessions (VTS with a Human Model in Fig. 2). After a short break, the “Mask-Making” activity was introduced, where students were presented with the artwork “The Sick Child” by Edvard Munch (1926) and asked to use a mask to express their feelings about it in the broader context of medical education. The session concluded with students describing their masks and engaging in self-reflection on the mask-making experience, emphasizing the thought and reflection required in both creating and understanding art [43, 44]. Figure 3 illustrates examples of the Visual Thinking Activities (VTA).

Fig. 2.

VTS with a human model

Fig. 3.

Describe and draw and mask making activities

Control group

The Control Group participated in four 2-hour training sessions focused on observational exercises. The educational content and activities were adapted from two primary sources: the “TRAINING THE EYE” project from Harvard University and the work of Boudreau et al. (2008).

During the training sessions, an interactive lecture approach was used to emphasize the importance of observation skills. Students engaged in a variety of observational exercises designed to enhance their abilities in this area. Feedback on the students’ performance in these exercises was provided based on the framework outlined by Klugman et al. (2015).

The specific observational activities included:

1. Optical Illusions Game: This activity is rooted in the field of perceptual psychology. Students explored a collection of optical illusion images, which demonstrated that perception can be influenced by expectations and biases rather than objective reality [1].

2. Blind Contour Drawing: This task is based on the concepts of motor learning and visual perception. Participants focused on carefully observing the contours of an object (a shoe) while attempting to draw it without looking at the paper. This exercise was aimed at improving hand-eye coordination and the ability to closely attend to details [45].

3. Focus on the Face: This exercise connects to Gestalt psychology. Working in pairs, students quickly sketched each other’s faces, concentrating on the relationship between different facial features and proportions. This activity encouraged a holistic approach to observation and representation [45].

For homework assignments, students engaged in two additional observational tasks:

1. Close Observation and Documentation: Participants observed an individual’s posture and quickly sketched the person’s pose, emphasizing balanced attention to details and the overall gestalt [45].

2. See Change: This activity is grounded in the principles of longitudinal observation and developmental psychology. Students acted as observers, reframing everyday occurrences (such as the healing of a wound) as opportunities to witness gradual change over time [1, 10].

After observational exercises, the students reflected on and discussed their experiences. These activities were designed to enhance the participants’ analytical, descriptive, and perceptual skills, as well as their ability to recognize and communicate observations effectively (Appendix C).

Measurements

The study design included a pre-test and post-test assessment to measure the effectiveness of the observational training program. The pre-test assessment consisted of a baseline survey to gather demographic information and details about the participants’ prior clinical or humanities education experience. Additionally, the pre-test included a written response section where students were presented with three art images and asked to describe their observations. These pre-test images were distinct from those used in the training course (see Appendix D). The selection of art images for the pre-test and post-test was guided by the educational design team’s expertise in visual thinking and informed by similar studies conducted in this area [18, 37]. The pre-test assessment was administered through an electronic polling system (ePoll), which allowed students to type their written responses. The post-test assessment was sent to all study participants, including both the intervention and control groups, one day after the sessions. The structure of the post-test was similar to the pre-test, with students again presented with art images and asked to provide written observations. The pre-test and post-test assessments utilized the same art images. To examine the intervention’s delayed follow-up effects and assess the extent to which students can apply the intervention’s outcomes in a real-world setting, an examination was conducted during an early clinical encounter session in the surgical department for both the intervention and control groups. The students were requested to articulate their observations from the clinical experience. This examination was also carried out using the ePoll system and was made available to both groups one day after the clinical exposure session. Differences in number of observations, word count, and time spent were calculated and compared between the intervention and control groups.

Validity and reliability of measurements

In studies examining the impact of VTS, the selection of appropriate images is a critical step to ensure the validity of the assessment measures. For this study, the instructional design team, who have expertise in VTS, was consulted to guide the image selection process. Initially, a pool of 16 candidate images was identified and thoroughly discussed among the expert team. Guided by established criteria for visual thinking and informed by similar studies in this domain [18, 37], the team reached a consensus on three images to be used for the pre-test and post-test assessments. Students were then prompted to provide written descriptions of these selected images, answering the question: “What do you see in this image?” To establish inter-rater reliability, two independent raters analyzed 20% of the student responses. The agreement coefficient between the raters’ scores was calculated as the inter-rater reliability coefficient. Any discrepancies between the two raters were resolved by consulting a third rater. The inter-rater reliability coefficient for the pre-test was 0.79, and for the post-test, it was 0.86. These values fall within the acceptable range for inter-rater reliability, indicating a high degree of consistency in the scoring of the student responses. Refer to Appendix E for an overview of the study.

Data analysis

To assess the written responses provided by the students, the number of words used in the free-response descriptions was calculated. Consistent with the methodology employed in previous studies [18, 40], the word count was performed using Microsoft Word 2019. Additionally, the total number of observations made by the students was determined by counting the individual factual declarations about the images. This process was carried out by one rater (N) and randomly verified by a second rater (M), with input from the instructional design team to define a single observation as “a single factual declaration about an image.” The time spent by the students in observing and analyzing the images was calculated by recording the time spent writing their observations and completing the assessment form within the ePoll software platform [18, 40]. To ensure the normality of the data, the study relied on the Central Limit Theorem. Considering that the sample sizes in the intervention (n = 44) and control (n = 45) groups were both greater than 30, and the samples were slightly discrete, the distribution of the means of this data was considered normal, regardless of the shape of the distribution in the target population [46–48]. In line with the study objectives, descriptive statistics, including mean (standard deviation) and frequency (percentage frequency), were used to characterize the data. To analyze the differences between the control and intervention groups, one-way ANOVA/ANCOVA was employed. Additionally, independent t-tests were conducted to compare the means between the two groups. All statistical tests were two-tailed, with an alpha of 0.05 considered the threshold for statistical significance. The data analysis was performed using SPSS software version 26.

Results

Participants characteristics

The study sample consisted of 89 first-year medical students, with 44 participants in the intervention group and 45 in the comparison group. An analysis of the socio-demographic characteristics revealed no statistically significant differences between the two groups, except for prior art activities. The mean age of the control group was 19.60 (SD = 1.19), while the intervention group had a mean age of 19.15 (SD = 1.02). The control group had 18 male participants, and the intervention group had 21 male participants. Additionally, 45 participants in the control group and 42 in the intervention group reported being single, with no statistically significant differences observed between the groups (p > 0.05). Regarding prior art activities, the control group had a higher proportion of students with such experience (31.5%, n = 28) compared to the intervention group (18%, n = 16). This difference was statistically significant (p = 0.01), as assessed by the question, “If you have prior art experience, please describe the type of activity.” To assess the participants’ humanities education, two questions were asked: “Please specify any formal university education experience you have participated in before entering medical school” and “Please specify your high school major.” None of the participants in the intervention or control groups reported having prior education in humanities, art, or clinical experience (Table 1).

Table 1.

Sociodemographic and training characteristics among intervention and control group first-year medical students participating in the study at the University of Gonabad (n = 89)

Variable		Control	Intervention	P-value
		Mean (SD) / Frequency (%)	Mean (SD) / Frequency (%)
Age		19.60 (1.19)	19.15 (1.02)	0.96
Gender	Male	18 (20.2)	21 (23.6)	0.52
	Female	27 (30.3)	23 (25.8)
Marital status	Single	45 (50.6)	42 (47.2)	0.24
	Married	0 (0)	2 (2.2)
High School Major	Empirical Science	43 (48.3)	44 (49.4)	0.49
	Mathematics	2 (2.2)	0 (0)
Prior Art Activity	No	17 (19.1)	28 (31.5)	0.01
	Yes	28 (31.5)	16 (18.0)
Previous Academic Major	No	44 (49.4)	43 (48.3)	0.99
	Yes	1 (1.1)	1 (1.1)

Short term assessment

A: Number of observations

The analysis revealed a statistically significant difference in the number of observations made by the participants between the intervention and control groups. Specifically, the participants who received the VTS intervention made a greater number of observations compared to the control group. The effect size of the VTS intervention on the number of observations was 0.12, as reported in Table 2; Fig. 4.

Table 2.

Number of observations among intervention and control group first-year medical students at the University of Gonabad (n = 89)

Model	Time point	Intervention n = 44 Mean (SD)	Control n = 45	Mean Difference (95% CI)	Partial Eta2	P value$
Crude	Pre	21.32 (14.57)	21.16 (14.56)	-	0.18	< 0.001
	Post	56.23 (37.95)	28.84 (16.87)	27.38 (15.05_39.71)
AdjustedA	pre	21.32 (14.57)	21.16 (14.56)	-	0.12	0.001
	Post	56.19 (37.61)	28.91 (16.64)	27.27 (15.57_38.97)

$ calculated based on ANOVA / ANCOVA Models

A: Adjusted for baseline Pre-score OUTCOME (calculated based on Oneway ANOVA / ANCOVA model)

Fig. 4.

Mean differences of the total number of observations in control and intervention groups

B: Number of words used to describe images

The results demonstrated a statistically significant difference in the number of words used by participants to describe the images between the intervention and control groups. Specifically, the participants who received the VTS intervention used a greater number of words to describe their observations compared to the control group. The effect size of the VTS intervention on the number of words used in the image descriptions was 0.21, as shown in Table 3; Fig. 5.

Table 3.

Number of words among intervention and control group first-year medical students at the University of Gonabad (n = 89)

Model	Time point	Intervention n = 44	Control n = 45	Mean Difference (95% CI)	Partial Eta2	P value$
Crude	Pre	108.75 (125.39)	94.38 (102.05)	-	0.17	< 0.001
	Post	265.48 (212.30)	115.44 (95.91)	150.03 (80.88_219.18)
AdjustedA	pre	108.75 (125.39)	94.38 (102.05)	-	0.21	0.001
	Post	264.89 (203.31)	121.06 (91.79)	143.83 (78.19_209.46)

$ calculated based on ANOVA / ANCOVA Models

A: Adjusted for baseline Pre-score OUTCOME (calculated based on Oneway ANOVA / ANCOVA model)

Fig. 5.

Mean differences of the total number of words in control and intervention groups

C: Time spent on art images observations

The results revealed a statistically significant difference in the time spent by participants on their observations between the intervention and control groups. Specifically, the participants who received the VTS-based intervention spent more time observing and analyzing the images compared to the control group. The effect size of the VTS intervention on the time spent on observations was 0.17, as reported in Table 4; Fig. 6.

Table 4.

Time spent on art images observations among intervention and control group first-year medical students at the University of Gonabad (n = 89)

Model	Time point	Intervention n = 44	Control n = 45	Mean Difference (95% CI)	Partial Eta2	P value$
Crude	Pre	14.00 (10.02)	13.42 (10.67)	-	0.14	< 0.001
	Post	21.07 (12.93)	11.53 (10.62)	9.54 (4.55_14.51)
AdjustedA	pre	14.00 (10.02)	13.42 (10.67)	-	0.17	< 0.001
	Post	20.99 (11.60)	11.72 (9.82)	9.26 (4.55_14.51)

$ calculated based on ANOVA / ANCOVA Models

A: Adjusted for baseline Pre-score OUTCOME (calculated based on Oneway ANOVA / ANCOVA model)

Fig. 6.

Mean differences of the time spent on describing observations in control and intervention groups

Delayed follow-up assessment

In the early clinical exposure test, the analysis of the total number of observations and words used by participants to describe their findings showed no statistically significant differences between the control and intervention groups. However, the independent t-test results revealed a statistically significant difference in the time spent by participants on the observational task between the two groups. Specifically, the participants who received the VTS -based intervention spent more time on their observations in the delayed follow-up compared to the control group. The effect size of the VTS intervention on the time spent during the primary clinical exposure test was 0.44, as reported in Table 5.

Table 5.

Observation skills of first-year medical students during early clinical cxposure in intervention vs Control Groups (n = 89)

Variable	Control	Intervention	Cohen’S D	P-value
	Mean (SD)	Mean (SD)
Number of words	154.31 (100.36)	165.48 (123.63)	0.09	0.64
Number of observations	25.93 (16.52)	28.00 (22.53)	0.105	0.62
Time spent on describing observations (in minutes)	12.33 (11.35)	17.80 (13.54)	0.438	0.04

Discussion

The primary objective of this study was to investigate the short-term and delayed follow-up impacts of a VTS intervention, grounded in the PRISM model, in comparison to a series of standard observation exercises on the observation skills of first-year medical students. Specifically, we assessed the total number of observations, the number of words used to describe those observations, and the time spent on the observational tasks.

In the immediate post-intervention assessment, the results demonstrated that the participants who received the VTS-based training made a significantly greater number of observations compared to the control group. This finding aligns with previous studies by Naghshineh et al. (2008) and Klugman et al. (2015), which reported that VTS-based interventions can enhance the overall number of observations made by learners when analyzing art and clinical images. Furthermore, our results corroborate the findings of Agarwal et al. (2020), who found that VTS-based approaches increased the total number of observations made by medical students when examining clinical images such as ECGs and radiographs.

Additionally, the VTS-trained participants used a significantly greater number of words to describe their observations compared to the control group in the short-term assessment. This outcome is consistent with the studies by Naghshineh et al. (2008) and Klugman et al. (2015), which indicated that VTS can lead to an increase in the verbalization and detailed description of observations made on art and clinical images. Similarly, Agarwal et al. (2020) demonstrated that VTS enhanced the number of words used by medical students to articulate their observations of clinical images.

another noteworthy short-term finding was that the participants who received the VTS spent significantly more time describing their observations compared to the control group. This result is in line with the study by Klugman et al. (2015), which suggested that VTS can increase the time spent by learners in describing art and clinical images. Similarly, Agarwal et al. (2020) found that VTS led to an increase in the amount of time medical students spent describing their observations of clinical images, such as ECGs and radiographs. However, in the study by Naghshineh et al. (2008), the authors were unable to assess the intervention’s impact on observation time.

Further analysis of the effect sizes revealed that the VTS had a moderate impact on the total number of observations made by the participants (effect size = 0.12), a strong impact on the number of words used to describe the observations (effect size = 0.21), and a strong impact on the time spent on the observational tasks (effect size = 0.17). These effect sizes suggest that the VTS grounded in the PRISM model had a meaningful and substantive impact on improving the observational skills of the first-year medical students in the short term [49, 50].

Observation, referred to as “seeing” in the context of art, is widely recognized as a non-objective process. Experts have argued that traditional medical education methods are insufficient in overcoming the constraints of conventional ways of seeing, and instead, a more flexible and dynamic approach, such as art-based training (e.g., VTS), is required [51]. However, the question remains: how can the full impact of engaging with visual artworks be accurately measured? A combination of quantitative and qualitative tools can help address this inquiry [8, 17, 26, 52–54].

The literature suggests that observation becomes more objective through the act of description. Individuals must be able to verbally or in writing articulate their visual experiences for the observation to be complete and meaningful. This concept of “deep seeing” involves repeated, accurate observations that are also emphasized in clinical settings, where learners are encouraged to see beyond their preconceptions when examining patients and interpreting medical images [1, 51]. Ike and Howell (2022) have proposed that quantitative assessments of observation skills should include the analysis of the number of distinct observations made, the number of words used to describe the visual stimuli, and the time spent on the observational tasks [17]. In the context of physical examinations and clinical diagnoses, observation is a crucial element, as physicians must observe, describe, and interpret visual information during the diagnostic process. Therefore, enhancing observational skills may lead to the generation of more accurate diagnostic hypotheses, resulting in improved and more cost-effective patient care. Ultimately, this would revive a tradition where diagnostic precision is not solely reliant on technological advancements, but rather on the clinician’s acumen and powers of observation [2, 6, 7].

An important finding of this study was that in the delayed follow-up, the two groups – those trained with the VTS intervention and those who received the Observation Exercise training – did not exhibit statistically significant differences in the total number of observations made or the number of words used during the post-primary clinical exposure test. This suggests that both interventions were equally effective in enhancing these specific observational skills in a real-world clinical setting over the delayed follow-up. However, a notable difference was observed in the time spent by the two groups in describing their clinical observations. The VTS-trained group spent significantly more time engaged in this descriptive task compared to the Observation Exercise group. The effect size calculations revealed a small effect size (0.44) for the time spent on observation, while the effect sizes for the number of observations (0.10) and number of words (0.09) were negligible [49, 50].

This finding indicates that the VTS had a more pronounced impact on the participants’ willingness to devote time and attention to the careful observation and description of clinical encounters, even if the ultimate quantity of observations and verbalization did not differ significantly between the groups. This aligns with the underlying principles of the VTS, which emphasizes the importance of deliberate, open-ended observation and detailed description as a means of developing deeper visual awareness and understanding [55]. It is worth noting that the previous similar studies did not investigate the delayed follow-up sustainability of the observed effects, making direct comparisons with the current findings challenging. In the present study, the follow-up period was limited to one-month post-intervention due to the constraints of the medical school curriculum, and only a single clinical session could be provided. These methodological limitations may have influenced the results and highlight the need for further research to explore the delayed follow-up impacts of such interventions, as well as the potential benefits of increased clinical exposure opportunities for students.

A key consideration in the implementation of Visual Thinking Strategies (VTS) is the need to ensure that the observed benefits in developing observational skills can be effectively transferred from the art-based context to clinical settings involving real patients. Previous studies have highlighted this as an important gap that requires further investigation [9, 18, 20, 40]. Recognizing this, the present study sought to provide a more clinically-relevant context for the VTS training by incorporating the use of a clinical skin image and a live human model in the fourth session of the intervention group. This allowed the students to directly apply the VTS approach to a clinical scenario, rather than solely focusing on artwork, as has been the case in many previous studies. By examining how the students engaged with VTS in this more authentic clinical context, the researchers were able to gain valuable insights into the participants’ ability to transfer and apply the observational skills developed through the VTS approach to real-world medical settings. Additionally, the inclusion of an early clinical exposure session as part of the delayed follow-up provided an opportunity to directly assess the sustained impact of VTS on observational skills during actual patient encounters. This multi-pronged approach is particularly important given the challenges inherent in following up with students in clinical settings, as well as the recognition that VTS may not be the only variable influencing student observational performance in such contexts [18]. By incorporating these clinically-focused components, the researchers have sought to address a key gap in the existing literature and better understand the real-world applicability and delayed follow-up effects of the VTS intervention. The findings from these clinical application and follow-up assessments will be crucial in determining the broader effectiveness and transferability of the VTS approach, ultimately informing efforts to optimize the development of observational skills among medical students. Continued research in this area, with a focus on longitudinal evaluation and the integration of VTS within diverse clinical settings, will be essential in advancing our understanding of the most impactful strategies for cultivating observational proficiency in future healthcare providers.

A noteworthy finding from the present study was that the intervention group, which received the Visual Thinking Strategies (VTS) training, spent more time observing and describing their observations compared to the control group, both in the short-term and delayed follow-up assessments. This is a significant result, as one of the key goals of VTS, as well as the practice of clinical medicine, is to instill in students the habit of spending considerable time carefully observing and articulating their observations [18, 40]. This approach to observation trains students to focus on the nuanced details of what they perceive, which is crucial for accurately observing patients and their clinical presentations. Such deliberate, time-intensive observation can lead to an “expansion of visual fields” and enhanced “situational awareness” – critical skills for medical professionals.

Particularly for pre-clinical medical students, the ability to spend time carefully observing real or simulated patients, and to focus on the findings of the physical examination, is an explicit component of developing observational competence. This “visual literacy” represents a form of aesthetic reasoning that emerges from the practice of careful observation, which can inform physicians about aspects of the patient’s condition that may not be explicitly communicated. By considering non-verbal cues in the patient’s facial expressions, emotions, body language, and background characteristics, the physician can construct a more holistic “picture” of the patient and their unique circumstances. This broader perspective, encompassing the emotional, psychological, and spiritual dimensions of the patient’s experience, is essential for providing comprehensive, patient-centered care [3, 12, 18, 40].

A critical factor influencing the outcomes of studies in this domain is the dosage of VTS training, which refers to the number and frequency of VTS instructional sessions [8, 18]. Previous research has indicated that the optimal dosage for VTS training remains unclear, as most existing interventions have been relatively brief, typically involving one to three sessions [8, 12, 18]. This suggests a need for further investigation to determine the most effective approach for delivering VTS training within medical education curricula [18]. In the present study, we selected a VTS training program comprising four educational sessions. This dosage of VTS training may have positively impacted both the short-term and delayed follow-up outcomes observed [12]. Additionally, this in-class VTS may be particularly beneficial in situations where access to art museums and specialized VTS instructors is limited [23, 41, 56].

The delayed follow-up data revealed that the VTS intervention resulted in the intervention group spending more time describing their observations than the control group. However, the total number of observations and words used appeared similar. These results imply that ongoing interventions and the incorporation of arts-based activities into medical curricula are essential for significant and lasting improvements in observational skills. This is supported by Ferrara et al. (2020), who found that repeated VTS interventions throughout medical training may be necessary to sustain benefits in visual skills. Additionally, Naghshineh et al. (2008) identified a dose-response relationship between the number of VTS sessions and observed effects. Haidt’s framework for arts-based education design and evaluation offers a solid theoretical basis for this progressive approach in medical education [9, 37, 57].

An important outcome of the present study was the finding that the VTS technique, when integrated with medical student interactions, could support the potential growth of observational skills in alignment with the first function of the Prism Model (i.e., mastering skills). This is a significant contribution, as previous studies have identified a notable gap in the use of theoretical frameworks to guide the implementation of arts-based programs, such as VTS, within medical education [8, 9, 26, 32, 58]. In this study, we designed and implemented a VTS program based on the Prism Model, drawing from Housen’s theory of aesthetic development, Arnheim’s visual thinking, and constructivist theory. The VTS method, emphasizing active learner participation, initiates deep, reflective thinking. Facilitators guide students to analyze and describe images through structured questions, expanding knowledge through personal exploration and fostering cognitive development. Learners’ outcomes are supported by existing evidence and cognitive synthesis of experience and new concepts [39, 59, 60].

The VTS experience, characterized by the co-construction of knowledge under facilitator guidance and peer collaboration, provides medical students with the Zone of Proximal Development (ZPD) necessary for growth, as conceptualized by Vygotsky’s theory [59, 60]. This approach can serve as an effective educational method in medical education, as it aligns with the principles of constructivist theories. The integration of VTS in medical curricula can help students develop observational and diagnostic skills, interpret complex meanings, and problem-solve in a non-clinical setting, thereby laying the groundwork for learning in real clinical environments [39, 60]. Furthermore, the Prism Model, akin to a prism, allows us to approach the targeted skill development from multiple perspectives, considering dimensions beyond just skill acquisition [27]. Haidet’s framework offers a structured approach to integrating arts-based education, which involves examining the unique qualities of the arts, learner engagement with art, short- and long-term learning outcomes, and ultimately the construction of new meanings and their translation into medical practice [9]. This comprehensive framework can justify the integration of arts-based programs, such as VTS, and promote meaningful learning.

Strengths and limitations

This study has several strengths that enhance the validity of its findings. First, the controlled group design with alternative training effectively isolates the specific effects of the VTS intervention. Second, integrating both VTS and control group interventions into the existing curriculum improved response rates and sample size, bolstering the generalizability of the outcomes. Third, the study addressed a key limitation from previous research regarding the transferability of VTS skills to clinical settings by incorporating a clinically relevant VTS session and a delayed follow-up evaluation on clinical skills application. Fourth, the intervention benefited from established educational frameworks, such as the AAMC PRISM model, guiding the integration of arts and humanities within the medical curriculum. The four-session VTS training aligns with emerging recommendations for optimal delivery in medical education.

However, there are limitations to consider. The study’s single-institution setting and relatively small participant pool may restrict the generalizability of findings. A larger, multi-institutional study is warranted for broader applicability. Additionally, while the design was controlled, it lacked randomization, which may introduce selection bias. Conducting the early clinical exposure on different days for each group could have affected the diversity of patient encounters. Lastly, participants were in the pre-clinical stage, which may limit their engagement with clinical settings compared to students in active patient care, impacting the transferability of skills developed through the program. Future research should focus on enhancing the generalizability of these findings through larger, multi-institutional, and possibly randomized studies.

Conclusion

This study suggests that incorporating VTS into preclinical medical education, along with the integration of conceptual frameworks such as the Prism Model can contribute to the improvement of medical students’ observational skills. The findings indicate that preclinical medical students trained using the VTS approach made more observations, used a greater number of words to describe art images, and demonstrated an increased tendency to spend more time on their observations. These students exhibited a more careful examination of the images, paying closer attention to details that they may have initially overlooked, allowing them to perceive elements that were not part of their initial observations. This technique appears to add depth to the students’ observations, teaching them to progress from more superficial or fewer observations to more in-depth and multifaceted perspectives.

With an understanding of how to effectively utilize art and the expected learning outcomes, the VTS technique, when interacting with carefully selected artworks, can support potential growth in observational skills, aligning with the first function of the Prism Model (mastering skill).

However, in the delayed follow-up and in a real-world setting, the study found that the intervention group only spent more time describing their clinical observations. This suggests that both the VTS interventions and the Observational Exercises in the control group were equally effective in the delayed follow-up and in a real-world environment in increasing the total number of observations and the number of words used for description. This finding, while significant, calls for further investigation in future studies, particularly those involving clinical rotations to validate and expand upon these results.

Electronic supplementary material

Below is the link to the electronic supplementary material.

Abbreviations

VTS

Visual Thinking Strategies

VTA

Visual Thinking Activity

OE

Observation Exercises

TUMS

Tehran University of Medical Sciences

GMU

Gonabad University of Medical Sciences

Author contributions

All authors contributed to the study’s conception and design. NGH and MA acquired and interpreted data. The first draft of the manuscript was written by NGH and MA. All authors commented on previous versions of the manuscript. All authors read and approved the final manuscript. The development and execution of the study were collaboratively conceived and designed by all authors. NG and MA were instrumental in the initial drafting and substantial revision of the manuscript. They, along with AK, undertook the acquisition of data, while AS provided critical analysis and interpretation of the data. FY offered essential consultation and contributed to the refinement of the final manuscript. Each author has actively engaged in the iterative review process, providing insightful edits and feedback. The manuscript has been read and approved by all authors, ensuring collective endorsement of the content presented.All authors contributed to the study’s design and manuscript development. NG and MA were central in drafting and revising the manuscript, as well as in data handling. AS analyzed the data, while FY provided key consultations. All authors reviewed, edited, and approved the final manuscript.

Funding

This project is funded by the Tehran University of Medical Sciences, School of Medicine, Tehran, Iran.

Data availability

The dataset supporting the conclusions of this article is included within the article (and its additional files). Due to the specific data-sharing regulations applicable to this study, the datasets generated and analyzed are not available in the public domain. However, they can be obtained from the corresponding author should there be a justified request.

Declarations

Ethics approval and consent to participate

The Ethical Committee of Tehran University Medical Sciences granted ethical clearance for the study (ethics code: IR.TUMS.MEDICINE.REC.1402.094). Study participants received comprehensive information regarding the research objectives, ethical guidelines, and their rights, leading to their informed and voluntary engagement. Consent was documented through participants’ signatures, affirming their understanding and agreement. Confidentiality and anonymity were upheld in the utilization of the data for scholarly discourse and publication. Participants retained the right to discontinue their involvement at any stage, with the assurance that all data and findings would be exclusively applied to research endeavors. Furthermore, the study’s methodology and execution adhered strictly to the Helsinki Declaration’s ethical standards.

Consent for publication

We obtained the consent from all participants to present their images.

Competing interests

The authors declare no competing interests.