Raju, the peon at my high-school in Mumbai growing up, lived (with a big family of 12) in a tin-roof shanty in nearby Dharavi, the world's largest slum. Electricity was scarcely available to them. With the decline in prices of Photovoltaic solar energy, it became worthwhile for Raju to buy a PV unit, so we collected donations at school and a set of 2 PV collectors and a battery for storage of the electricity - so that electricity generated during the day could be consumed in the evening - were installed.

But things appeared to be a bit more compplicated than we had bargained for! After only 2 years the battery was not functioning as it should; it did not store as much electricity as it used to (and was designed for). The lights in the evening could run only upto about 10:30pm - whereas the manual promised an all-nighter! Raju complained about this to the manufacturer, but they replied that they were NOT to blame and it was Raju's fault (and his family's, by extension)!

When I heard about the issue, I visited with the family and after a bit of investigation discerned the problem. See, slum-dwellers in Dharavi were so used to scarcity in availibility of electricity that they were used to keeping the lights low and dim in order to be as frugal as possible when using electricity. But this 'saving behavior' (although perfectly rational) appeared to ruin the battery. How? To function properly, the battery NEEDED to discharge by giving a steady and strong current of electricity. Raju had unintentionally destroyed the new battery because of old habits that were persistent and did not fit new technologies!

This intriguing interaction between behavior and technology and its consequences are insightful for future design of technology AND policy!

How can we describe and understand these interactions?
What kind of scientific disciplines can add to these understandings?
What kind of empirical research has been done on this subject?
How can and should we deal with these interactions in the design of technology products?
Can we benefit from knowledge of these interactions for design of policy?