Research on Cognitive Domain in Geoscience Courses: Temporal and Spatial Reasoning (Cog A)

Post a Comment »

Authors: Katherine Ryker, University of South Carolina; Scott Brande, University of Alabama at Birmingham; Mariana Guereque, The University of Texas at El Paso; Alison Jaeger, Temple University, and Julie Libarkin, Michigan State University

Jump Down To: Grand Challenge 1 | Grand Challenge 2 | Grand Challenge 3 |

Introduction

The geosciences are characterized by their particular application of and reliance on temporal and spatial reasoning. Geoscientists must be able to apply their knowledge across a variety of scales. In the words of Arthur Conan Doyle, "From a drop of water, a logician could infer the possibility of an Atlantic or a Niagara without having seen or heard of one or the other. So all life is a great chain, the nature of which is known whenever we are shown a single link of it." Geoscientists should be able to look at, say, physical and chemical differences in sedimentary layers from a core of the seafloor and infer changes in erosional patterns (spatial) over time (temporal). The ability to engage with this kind of task represents a great shift in thinking from where most students begin their studies, be that in K-12 or college. In order to understand how people's ability to spatial and temporal reasoning how that thinking changes over time requires

While more is known about how people reason spatially as compared with temporally, there are still significant gaps in our understanding of spatial reasoning in the geosciences. We believe that there are opportunities to build on lessons learned from previous investigations of spatial thinking, including how a community can investigate a specific line of reasoning (e.g. the Spatial Intelligence and Learning Center, or SILC). There is also a need to build on established research from other domains, including cognitive psychology and anthropology.

We identified three Grand Challenges to better understanding spatial and temporal reasoning. These include identifying what reasonings or skills are essential to the geosciences (both broadly and within subdisciplines), and the intertwined challenge of how to assess those reasonings and use those results to improve on what students are learning from their geoscience experiences. We also see an additional grand challenge as being the need to connect with and build on the work done by other disciplines and communities.

Jump Down to: Grand Challenge 1 | Grand Challenge 2 | Grand Challenge 3 | Grand Challenge 4

Grand Challenge 1: What are the types of spatial and temporal reasoning essential to the geosciences, and where do research gaps remain within this domain?

Rationale

In order to make sure our work is relevant to the broad geoscience education community, we first need to understand what the key spatial and temporal tasks that we want students to be able to tackle are to that community. Essential types of spatial and temporal reasonings need to be "mapped" so that we know where the gaps are.

Research Strategies

  • Kastens & Manduca (2012) created concept maps of Spatial Thinking and Temporal Thinking in Geosciences (below). These should be revisited and updated. One idea to illustrate where gaps exist in our understanding would be to make these concept maps interactive. For example, node color could indicate the thoroughness of study and clicking on each node would bring up citations for relevant studies. However this update is done, revised maps of the landscape will help us identify strengths and gaps and identify dimensions on which mapping needs to happen.

  • Researchers should consider whether there are different essentials reasonings for different aspects of the geosciences, for example remote sensing, field work, and mineralogy. These essential reasonings are the different skills and abilities that help people function in their roles.
  • Select specific, well-defined areas of geoscience and have people in those fields describe the spatial and temporal tasks they do as part of their job in focus groups. We recommend that focus groups might help elicit more ideas than one-on-one interviews or surveys. This cognitive task analysis with specific experts could be used to identify the most important, or essential, spatial and temporal reasoning tasks they do. This could also be completed as a modified Delphi study, or by studying geoscientists doing expert tasks, and coding for different reasonings being used.

Grand Challenge 2: How can geoscience education foster spatial and temporal reasoning, and how can assessment measure spatial and temporal reasoning?

Rationale

With an understanding of the essential types of spatial and temporal reasoning for the geosciences, we can then proceed to assess how these reasonings change over time in response to different learning experiences. We want to know how best to teach students to accomplish tasks that rely on these essential reasonings. This leads to the development of two intertwined challenges: assessing those reasonings and developing strategies that foster them. This is not to suggest that an exhaustive list of reasonings must first be developed before these two intertwined Grand Challenges can be addressed. Rather, researchers should first establish that the particular reasoning they are studying is critical to some aspect of success in the geosciences. Following that, it is imperative that the correct measurement tools are used. Many tools already exist, especially to assess spatial thinking (see spatiallearning.org for some examples), while others likely need to be developed.

Research Strategies

  • A comprehensive literature review would be of great benefit in establishing what assessment tools already exist and what they measure.
  • Researchers might have students complete sorting tasks (e.g. in order of size or amount of time) to better understand at what scales that reasoning begins to fall apart.
  • Recognize that learning requires students to be engaged, select relevant information, organize that information, and integrate it with prior knowledge. Identify features of different instructional techniques (including timing) that impact each of these stages, and whether they impact one phase more than another.
  • Both novices and experts utilize their intuition in assessing spatial and temporal problems, but novices can suffer from inappropriate or "over-zealous" transfer. Researchers can identify explicit models that novices and experts are relying on to do these kinds of thinking.
  • Study transferability from general, content-agnostic skills to discipline-specific skills and possibly vice-versa. Does training in a content-agnostic skill influence the development of a discipline-specific skill in any way?
  • Develop studies that get at why certain techniques work better than another.
  • Identifying or developing additional metrics to assess the spatial and cognitive structure of geoscience tasks.
  • Mapping learning progressions for tasks involving either spatial or temporal reasoning.
  • To what extent is transfer possible between content-agnostic and domain-specific tasks? For example, analyzing patterns of erosion and deposition in a stream requires an application of the broad concept of vectors; studying patterns of climate change requires students to do work in both spatial (scaling up from a local to global level) and temporal domains (change over time).

Grand Challenge 3: How can we reach out to other domains/communities to explore spatial and temporal reasoning?

Rationale

We recognize that the geosciences are not the only field interested in developing and using spatial and temporal reasoning. We can and should draw on other fields, including cognitive science and psychometrics. Other DBER areas, like Physics Education Research, have already done significant groundwork that could be built upon.

Research Strategies

  • Research on spatial and temporal thinking in the geosciences should be informed by other fields and the results of their research. To do this, we need to identify domains, communities and disciplines most likely to have knowledge related to the other Grand Challenges and connect with them. Some examples of relevant communities include cognitive psychology, educational psychology, learning sciences, linguistics, AI, education, communication, economics, anthropology, ethnography, cognitive neuroscience, evolutionary biology, art and design, data/media literacy, and sociology. In order to identify what domain to connect with, it is likely that researchers need to think carefully about a content-agnostic equivalent of the task in which they are trying to train students.
  • Identify potentially relevant Science of Learning centers.
  • Host something like a Gordon research conference to create interdisciplinary conference around spatial and temporal reasoning. This would help bring currently disparate researchers together around a central theme.
  • There are far fewer studies of temporal reasoning than spatial reasoning. This represents a large area for research to develop and includes issues of scale and rates.




Cognitive - Spatial and Temporal Reasoning -- Discussion  

Excellent summary of the challenges and research strategies we can use to tackle them. I am particularly interested in the problem of measuring spatial and temporal thinking. In addition to cataloguing what assessment tools already exist, we need to know which ones have been validated and shown to be reliable. Those that have not yet been validated need to be. Another potential research strategy is collecting and analyzing errors in spatial and temporal thinking - these can provide us with insights into where we ought to focus instructional efforts.

12417:34649

Share edittextuser=541 post_id=34649 initial_post_id=0 thread_id=12417

Originally Posted by Carol Ormand


Excellent summary of the challenges and research strategies we can use to tackle them. I am particularly interested in the problem of measuring spatial and temporal thinking. In addition to cataloguing what assessment tools already exist, we need to know which ones have been validated and shown to be reliable. Those that have not yet been validated need to be. Another potential research strategy is collecting and analyzing errors in spatial and temporal thinking - these can provide us with insights into where we ought to focus instructional efforts.



Hi Carol, I'd be interested in your thoughts about how to validate measures of spatial and temporal thinking in geoscience education. I'm OK with reliability, in the sense of repeatability, but puzzled by validity. How do we develop confidence that we are measuring what we think/hope we are measuring? In satellite or airborne remote sensing, we have the process of ground-truthing; what is the equivalent for measures of spatial or temporal thinking in geosciences? Kim

12417:34695

Share edittextuser=75 post_id=34695 initial_post_id=0 thread_id=12417

Originally Posted by Kim Kastens



Hi Carol, I'd be interested in your thoughts about how to validate measures of spatial and temporal thinking in geoscience education. I'm OK with reliability, in the sense of repeatability, but puzzled by validity. How do we develop confidence that we are measuring what we think/hope we are measuring? In satellite or airborne remote sensing, we have the process of ground-truthing; what is the equivalent for measures of spatial or temporal thinking in geosciences? Kim



Hi Kim,

Great question, probably better answered by a psychometrician... But it seems to me that we can at least strengthen our case for validity through think-aloud interviews with test-takers, along the lines of what Kali and Orion reported in their 1996 paper.

12417:34704

Share edittextuser=541 post_id=34704 initial_post_id=0 thread_id=12417

This is a wonderful collection of next steps for the community. A few comments/additions:
GC#1: The maps by Kastens & Manduca have been very helpful to shape my thinking. I see some opportunity to cross walk some of this with the "conceptions" GCs. Specifically, there seems to be scant existing literature on spatial/temporal reasoning and surface processes. The challenges here also relate to "systems thinking" which I think is currently somewhat poorly defined - but could be the realm within which various spatial and temporal reasoning takes place.

GC#2: I tend to agree with Kim about the concern over valid measures of spatial thinking. The existing psychometric measures of spatial thinking individually or collectively predict something, but they do correlate with general intelligence, so what they are predicting can be misleading. It's not clear how each measure of spatial thinking relates to each other and more importantly to domain tasks. So perhaps this is more a note of caution: (1) We must involve psychometric experts to be sure not to over interpret our findings, and(2)There is potential for a rabbit hole of assessment generation. So perhaps this gets back to GC #1: having focus groups or delphi studies that identify "threshold" spatial thinking problems for undergraduates. What holds students back from persisting - because people are likely going to self-select sub-disciplines that lie within their spatial wheelhouse.

GC#3: Love these Research Strategies. Consider adding even more modest bullets: convening transdisciplinary workshops at locations that would engage diverse researchers in tackling one branch of the spatial/temporal reasoning maps.

12417:34707

Share edittextuser=16887 post_id=34707 initial_post_id=0 thread_id=12417

The research strategies given for approaching these Grand Challenges are exciting. My interest is especially piqued by research strategies for Grand Challenge #1. I like the idea of revisiting the spatial and temporal thinking concept maps to identify strengths and gaps and I see some overlap with Grand Challenge #3, particularly with recognizing spatial/temporal reasoning that has been identified in other domains. Revisiting alternative spatial thinking frameworks and typologies to look for overlap and gaps could be productive. Consolidating research via interactive concept maps with nodes that indicate thoroughness of study would provide a great tool for furthering research.

12417:34815

Share edittextuser=22924 post_id=34815 initial_post_id=0 thread_id=12417

This post was edited by Kristen St. John on Jan, 2018
1. I’d like to see expanded and well-cited(!) rationales for each GC. This is the only WG that placed ALL of the references that we had collected last spring/summer into their draft chapter, but these are completely disconnected from the text and it looks hurried (and maybe it just was). So now is the time to work on making stronger, longer, and better supported rationales. Each GC should have its own set of references and those should only include citations that are made in that section. That said, I also like the idea of making all of the references we collected for a theme available publically. Perhaps that could be as an added Recommended Reading list or as Additional Resources list? But that isn’t a substitute for integrating specific key references into the intro, rational and recommended strategies for your theme. Please look at other WGs to see how they approached it. A good example is WG#2.

2. This relates to #1 above - Please delved into the literature (including the literature that you have in your list at the end) to make sure you are giving credit to the good work that has already been done. The Framework we are trying to construct needs to build on what has already been done. Here is one example: In your GC#3: you recommend a Gordon research conference. That might be good but only if it builds on what has already been done. There have already been at least two Gordon conferences on spatial thinking/visualization: one on Grand Challenges in the Use of Visualization in Science and Education in 20175 hosted by Bob Kolvoord and Katherina Scheiter and then in 2017 there was the Gordon conference on Scientific Visualization for Decision-Making that Mike Stieff and Ann Batiza ran and that Kim Kastens was a vice chair on. What came out of those conferences that helps establish what we know in spatial reasoning and what questions remain or new questions that arise from that work?

3. WG#3 included some suggestions for important researchable questions under each of their Grand Challenges, in addition to their recommended strategies. That wasn’t a requirement in the format, but I think it is very effective. Please look at WG3’s draft chapter and see if you think that is a good way to go to help give more concrete examples of important research directions in spatial and temporal reasoning in the geosciences.

4. There are 3 GC listed but links to 4 GCs (near the top) I know this was because some GCs were combined but that needs cleaning up.

12417:34854

Share edittextuser=1143 post_id=34854 initial_post_id=0 thread_id=12417

I also think the GC #2 strategy to identify current tools is a good idea, especially since many tools likely exist from a variety of disciplines. GC #2 includes a really interesting and important this of research topics, but without an accounting of the current state of assessment tools, it seems like they would be difficult to tackle. In gathering available tools, it would be particularly helpful to the GER community to identify those that are most promising, even if not yet validated, to help focus efforts. Conversely, perhaps the list of research topics could help inform the types of tools that are necessary to investigate such learning, followed by an investigation of whether or not a tool exists already. So I guess I'm not sure how/if GC #2 should be modified, but I think it is important and interesting!

12417:35157

Share edittextuser=22342 post_id=35157 initial_post_id=0 thread_id=12417

This post was edited by Kristen St. John on Feb, 2018
Here is an article that just came out that looks really valuable to this theme: https://eos.org/features/learning-to-form-accurate-mental-models.

12417:35175

Share edittextuser=1143 post_id=35175 initial_post_id=0 thread_id=12417

Really great discussion and set of well-articulated grand challenges. Picking up on many of the themes in this comment thread and especially bringing it to Kim's initial mention of validity, I am also often stuck here in research. It is clear that many of the items that pscyhometricians assert they are measuring is actually not trivial to map to geological reasoning. We have experience with the hypothesized connection from instrument to attempted measure not working, and in other cases the connections do show up, but have unexpected overlap with other measures. This to me means that we really don't understand (often, not always) what we are measuring in terms of geological spatial/temporal reasoning, even though we think we know what we are measuring psychometrically. This makes the validity question crucial, and I think opens really interesting windows into spatial reasoning in the geosciences in general. It potentially also offers interesting feedback to the psychometrics community. Much work to be done here for sure.

12417:35379

Share edittextuser=429 post_id=35379 initial_post_id=0 thread_id=12417

I like the GCs and strategies described here, I wonder about new technologies that are in development that will be used to represent earth processes that students will be able, and need to use, as well as new technologies that are in development that can be used in teaching to improve spatial reasoning.

12417:35394

Share edittextuser=749 post_id=35394 initial_post_id=0 thread_id=12417

I like the way you broke down the GC and strategies. I wonder if you may consider adding the development of easy to implement trainings and assessment strategies that are relevant to geoscience ed? Having a set of short trainings activities that could be used for remedial training with students with lower than average spatial skills may be helpful for geoscience departments. Those trainings could be abstract in nature (like many of the spatial skills tests are) to avoid barriers for students with limited geoscience content knowledge but may also include some exercises that include the transfer of spatial skill training into geoscience education.

12417:35415

Share edittextuser=3345 post_id=35415 initial_post_id=0 thread_id=12417

Join the Discussion


Log in to reply