Embedded Phenomena: Increasing comprehension of STEM concepts using body and space

Embedded Phenomena: Increasing comprehension of STEM concepts using body and space

By Carolina Kuepper-Tetzel

Two weeks ago, I went to the conference of the European Society for Cognitive Psychology (ESCoP) in Potsdam, Germany. One of the talks that I attended intrigued me so much that I decided there and then that I needed to write a blog post about it. The talk was presented by Dr Allison Jaeger, who is currently a postdoctoral fellow at Temple University, USA. In her research she investigates the benefits of students using their body and the classroom space to better understand complex concepts in science, technology, engineering, and mathematics (STEM).

In more abstract and philosophical terms: This research area looks into how bodily states and actions influence cognition (e.g., the way we think, remember, and comprehend) and vice versa. In simpler terms: How does what we do influence how we think and the other way around? In cognitive science, the idea that our body and mind are inevitably connected is summarized under the terms embodiment or grounded cognition (1). If we accept this idea, then we may be able to simulate complex phenomena with our bodies, actions, and the space around us to achieve a better understanding of complex and more abstract concepts. This may be particularly useful for disciplines such as STEM, for which spatial reasoning ability has been identified as the most important predictor for learning success (2). Spatial reasoning ability is “the skill required to orient or perceive one's body in space or to detect or reason about relationships within or between objects in space(3). Students vary in terms of their spatial ability skills and this can be one source of explanation why some students perform better than others in STEM. However, Dr Jaeger and her colleagues wanted to find out whether an embodiment intervention could benefit particularly students with lower spatial skills. Put differently, could giving lower spatial skilled students the opportunity to act out and directly experience a scientific phenomenon increase their comprehension of the concepts because they would not have to work this out mentally? In order to test this, they introduced Embedded Phenomena (4) as a classroom learning activity which uses the classroom space and encourages students to directly experience and simulate a scientific phenomenon with their bodies.

Jaeger et al. (5) conducted a 6-week field experiment in two 5th grade classrooms. One classroom was assigned to the embedded condition and the other one to the non-embedded condition. All students learned about earth’s layers, tectonic plates, and the development of earthquakes during Science class. During learning activities, students had to analyse seismograph data to calculate the location of the epicentre of the 15 simulated earthquakes. Here is where the two conditions differed from each other:

Image by Allison Jaeger

Image by Allison Jaeger

  • In the embedded condition, “RoomQuake” was used as Embedded Phenomena where pupils experienced simulated earthquakes in the classroom. Seismic reading stations were positioned at different locations in the classroom. When an earthquake started, pupils had to move around, collect data from these different seismic reading stations, and determine the position of the epicenter. These seismic reading stations could be assessed all day – thus, not only when an earthquake was happening. Students learned trilateration (NB: I just found out that this is the same method that GPS uses to determine a specific location on earth) by using strings to determine the location of the epicenter. Taken together, in the embedded condition, students would use the classroom space, their bodies, and strings to obtain information on the 15 earthquakes that happened at random times during a 4-week period.

Pupils in the embedded condition working together and moving around to locate the epicenter of the simulated earthquake. Video with permission of Allison Jaeger. 

Image by Allison Jaeger

Image by Allison Jaeger

  • In the non-embedded condition, historical earthquake data was used and students had to determine the epicentres of 15 earthquakes using trilateration on a map. In addition, the activity was not distributed across 4 weeks, but rather was completed within one week.

Using this design, the researchers wanted to explore two questions:

Q1: Do embedded activities increase learning outcomes?

Rationale for Q1: Direct experience of the phenomenon is a concrete instantiation of the abstract concept of calculating and analysing data. Consequently, the embedded condition is likely to increase understanding of the underlying processes and, in turn, result in better test scores on the final test.

Q2: Do embedded activities reduce the need for mental simulation? Therefore, embedded activities will particularly support students with low spatial reasoning skills.

Rationale for Q2: Outsourcing mental simulation to real-time simulations that pupils experience, likely decreases the burden to try to mentally simulate such large-scale natural phenomena. In the embedded condition, spatial cues from the room can be directly used. Consequently, if embedded activities reduce the need for difficult mental simulations, then using this intervention will be especially supportive for students with poorer spatial reasoning skills because students in the embedded condition would not need to rely as much on doing mental simulations.

Image by Allison Jaeger

Image by Allison Jaeger

Here are the results of their research endeavour: First, they find that pupils in the embedded condition outperformed students in the non-embedded condition on key learning measures such as comprehension of concepts, performance of analytic methods, and acquisition of seismological skills. Thus, introducing embodiment as a classroom activity shows a clear advantage for learning performance.


However, what I thought was even more intriguing are the effects of embedded instruction if looking at pupils with different spatial reasoning skill levels. If you look at the same learning performance measures again, but add the layer of spatial skills you will see that embedding real-time experience was particularly helpful for pupils with low spatial skills. In fact, in the embedded condition, students on all spatial reasoning skill levels perform equally well on all variables, i.e., comprehension of concepts, performance of analytic methods, and acquisition of seismological skills. (In scientific terms – and for people who know stats: Spatial reasoning skill was only a significant predictor for learning performance in the non-embedded condition, but not in the embedded condition.)


Although the authors point out limitations of their study such as small sample size and the need to replicate the findings with other Embedded Phenomena, I find that the current results have important practical implications for classroom instruction in STEM. If teachers are willing and have time to learn more about how to implement Embedded Phenomena, this can be a key for inclusive instruction. The findings that embodiment interventions seem to support student with poorer spatial reasoning skills is promising and should be taken into consideration when planning classroom activities.


(1)  Barsalou, L. W. (2008). Grounded Cognition. Annual Review of Psychology, 59, 617-645.

(2)  Wai, J., Lubinski, D., & Benbow, C. P. (2009). Spatial ability for STEM domains: Aligning over 50 years of cumulative psychological knowledge solidifies its importance. Journal of Educational Psychology, 101, 817-835.

(3)  American Psychological Association (2015). APA Dictionary of Psychology (2ed). Washington DC.

(4)  Moher, T. (2006). Embedded Phenomena: Supporting science learning with classroom-sized distributed simulations. In Proceedings ACM Conference on Human Factors in Computing Systems (pp. 691–700). New York: ACM Press.

(5)  Jaeger, A. J., Wiley, J., & Moher, T. (2016). Leveling the playing field: Grounding learning with embedded simulations in geoscience. Cognitive Research: Principles and Implications, 1, 23.