Do Life-Logging Technologies Support Memory for the Past? An Experimental Study Using SenseCam.
Abigail Sellen, Mike Aitken, Steve Hodges
“Experimentally evaluates the efficacy of still images in triggering the remembering of past personal events, having implications for how we conceive of and the claims we make about ‘life-logging’ technologies.”
A SenseCam hangs around your neck and captures many, many things.
Do we really want to remember everything (isn’t it important to forget)?
What is digital memory?
What is capturing an experience?
Can a machine remember?
Do life-logging systems in fact support human memory?
In what ways?
How does this change over time?
What kinds of data should we capture, and how?
There are alot of assumptions about the answers to these questions. There is only weak evidence that videos, still images and audio can trigger recollection of everyday events.
Relevant psychological concepts:
General knowledge (Semantic) vs. Recollection (Episodic) memory
Q1. Do Sensecam images actually improve our memory for past personal events?
Q2. In what ways do Sense cam images help us connect wiht our past?
Q3. How do cues strenghten or decline over time?
Q4. Are actively captured Sensecam images better retreival cue than similarly passively captured images.
– Lab control outside the lab
– 3 variables, 2 were
Condition: Sensecam or Control images
Interval: Short vs. Long
For one day a friend wore a sensecam. For two days, the participant wore their own sensecam. The next day, the friend wore it. The friend’s cam pics were considered control images. Addendum: These two are not actually friends and didn’t hang around each other, the were just considered experimental pairs. Later, the participant was given a free recall test to list all the events they can remember and the details of who-what-where-when. They distinguished between mentally recalling/reliving the event and a general knowledge memory of the event (“I usually eat lunch at this time”). They were given a recognition test to see an image and guess whether it was their sensecam image or not. And again, four months later for recall and recognition tests.
People remembered more on the days they wore the sensecam. This shows a novelty effect and a flaw in the experiment (they students should have worn a inoperative camera on the control days).
If you view your sensecam image you’re more likely to recognize the event. With your sensecam image you’re able to know a little more about what you’re up to even if you can’t remember them. They were good at distinguishing their own images rather than those of others.
Sensecam images improve people memory of the past. The images are meaningful to people on a personal level. Participants are better at ordering their own images than other people’s images. They might be looking at the images and recollecting the events. They could be using their own general knowledge of their routines to make deductions. The difficulty of ordering other people’s images suggests a personal aspect to this recollection. However, (this seems interesting) as people “generally forget”, this personal-advantage-in-sorting-your-own-images may be lost over long-term.
Claim: This stuff helps us with with information retrieval, but maybe not in remembering.
Taking an active picture was a worse trigger of memory? Because they were already likely to remember the moment anyway? I may have heard this wrong.
Did they not want to wear the Sensecam?
We still need to look at whether a fisheye 1st person photo is better than a 3rd person photo for triggering memories (as some have claimed).
Sharing aspect will be important.