Clickstream

In this post, we explore data collected with Anabasis webapp. 144674 records have been retrieved consisting of date, picture number, cell position and user id. We derive density plots from data to deduce where players prefer to click. Afterwards, we show how players share the work together on some selected pictures.

density inside room (top 5000 for each draw)

Where players prefer to click in the whole area?

For each drawing, we collect the first 5000 cells discovered and obtain density plot of cells:

density of clicking on the whole area

If cells were selected uniformly, each one should have density of 1e-4. We observe that densities for cells in the top-left room are mostly over 2e-4. The rooms located in the first row have many cells with density over 1e-4, with a progressive fading over columns. Middle-left (5th) column has higher densities compared to middle-right (6th) column. Concentric circles from the center with decaying densites seem appearent. Finally, cells located in rooms in middle-left and middle-right have lower densities (under 1e-4), especially for column 1, 2 and 9.

Where players prefer to click click inside a room?

For each drawing, we collect the first cells discovered. We obtain density plots of each cell inside a room. If cells were selected uniformly, each should have density of 0.01.

density inside room (top 5 for each draw)

Legend: Density plot inside a room where data are the first 5 cells of each drawing.

When a player arrives in a room, there is a strong preference for selecting top-left cell, and then drawing a diagonal towards the center.

density inside room (top 100 for each draw)

Legend: Density plot inside a room where data are the first 100 cells of each drawing.

After 100 cells discovered (among 10000), density concentrates on the center cells and on diagonals.

density inside room (top 5000 for each draw)

Legend: Density plot inside a room where data are the first 5000 cells of each drawing.

When many cells have been discovered, concentric circles from the center with decaying densites clearly appear, with a preference for diagonals and the middle cross cells (columns and rows 5 and 6).

How players share the work

The following pictures show four draws where one color is assigned to (almost) one id. I let you observe them.

draw 1 with one id for one color

draw 9 with one id for one color

draw 15 with one id for one color

draw 29 with one id for one color

I hope you have enjoyed this content! If you decide to modify or build a new app from my code, please let me in touch!

Related contents:

  • Some players informed me there was a game from Peter Molyneux with a related concept and called Curiosity – What's Inside the Cube?
  • Plots in this post have been done with ggplot, code is available here.
Written on June 11, 2015