Using visualizations in netlogo to enhance attention

Hi there,

I was hoping to start a conversation about adding visualizations to netlogo models to help people notice certain patterns and events in the model.

In my work designing models for learning units on various topics, we have found that some learners find netlogo models too visually messy to be able to notice things. As a result, we have developed several types of visualizations to help learners notice various aspects of models’ behaviors.

As an example, I’m attaching the wolf sheep predation model with three added visualizations, on the right side of the world:

  1. Focus on individual entities - you can follow a sheep or wolf until they die. This helps noticing individual’s behaviors by putting a halo on one of them.

  2. Focus on types of events - you can look at deaths and births of individual animals. This helps connect two patterns - individual life cycles, local contexts where these lives are impacted, and global patterns of population levels that are oscillating. For example, you may notice that there are more wolf deaths where there is a scarcity of sheep, and sometimes this is related to a downwards dip in population for wolves. More sheep are born where there is more grass and this is sometimes related to an upwards shift in sheep population.

  3. Including users’ actions into the model - you can join the model by hunting wolves, sheep or grass and examine the impact on the food chain/network. This type of action/visualization is very effective in focusing learners’ attention on local events in the model.

I was wondering whether any of you folks have been using such visualizations, and what you find their impacts are on learning. I can see this topic developing into some taxonomy of visualizations.

Sharona

You need to download the model from this page and then you can use it:
wolf sheep predation with visualizations

1 Like

Hi Sharona,

This is a cool topic! I have done something very similar with a model of a PN junction (attached). This is an electronic device made of two semiconductors, one “P-type” and one “N-type” and it forms the basis of all modern computer hardware (transistors are made of combinations of PN junctions). The n-type side has free electrons. The p-type side has “holes” which are empty electronic states conceptualized as positive particles in their own right. When an electron and hole meet, they “annihilate” (the electron fills the hole). In the model, I have two visualizations similar to yours:

  • there is a button to watch an individual charge carrier
  • a star appears whenever an electron and hole annihilate. These stars can be set to erase each tick (erase-stars?) or not if you want to see over time where they are happening (which reveals that most happen near the interface).

p-n-junction.nlogo (32.2 KB)

A couple of questions for you, regarding your model:

  • Did you consider distinguishing between sheep deaths from lack of food vs being eaten by a wolf?
  • Have you even worked with students to think about questions they have for the model and develop their own visualizations? This could be an interesting way to start students on modeling.

I’m mentioning this citation because it’s relevant to the general topic of visualizations:

Kornhauser D, Wilensky U, Rand W (2009) Design guidelines for agent based model visualization. Journal of Artificial Societies and Social Simulation 12: 1

I downloaded the model but it needs an include file, current-utils.nls. Where do we get that?

Jim

Whoops, sorry

@JimLyons Whoops, sorry about that. I just attached a version with all the code included.