The potential for search to augment authentic inquiry for children

In 1919, John Dewey described a mechanical pedagogy, one which encourages students to separate body and mind, recognition from meaning. In essence, Dewey was describing an education bereft of authentic inquiry, which I define (in agreement with Dewey) as the interminable process of exploring questions relevant to one's own existence to the end of uncovering ever greater, albeit incomplete, understanding. Nearly a century later, the mechanical pedagogy still looms large, even in an information-rich age where yottabytes of information are available to any student with access to an internet browser and a search box. Despite the power of the search, which may one day replace the lever of Archimedes as the tool by which one might move the world, the observation of Dewey holds fast: “we are very easily trained to be content with a minimum in meaning, and fail to note how restricted is our perception of the relations which confer significance” (Dewey, p. 6). Too often we are content to settle for the first query result that provides a reasonable explanation to our question, rather than probing deeper to refine and extend our original inferences. Behind the false assurances of the query result page, our doubt is reduced, and, likewise, our ability to think, as Dewey defines it.

In their investigations of how children search the internet with keyword interfaces, Druin et al reveal several of the aspects of contemporary pedagogy, and design features of prominent search engines themselves, that reinforce mechanical operations at the expense of thinking (Druin et al, 2010). Authentic inquiry for children is immediately inhibited by an interface that doesn't mesh with their typical strategy of spoken, natural language questions, which primary grade teachers and parents have become adept at deciphering. The fact that search engines struggle to match ostensibly cryptic, exploratory questions with database-driven, indexical search terms, or phonetic misrepresentations of words, such as “scedwal” for “schedule,” with their alternative spellings for common, adult typographic errors, imposes a system on children with which they are unfamiliar. The danger of this mismatch is an inhibition of inquiry. When faced with unsuccessful search results, Druin et al found that children immediately assumed that the information they sought was not available on the web. Strikingly, this was not the case in earlier studies, where children were given search tasks within finite datasets. On the contrary, subjects in these studies rigorously pursued their tasks when they encountered dead ends (Druin et al, 2010). The authors suggest a modern-day loss of confidence in technology, but what is more likely is a paucity of the spirit of authentic inquiry in our school-aged students. What students have learned, is pseudo-inquiry, the process of searching for a known answer in a limited data set, which is replicated time and again in formal education environments. When searching for an unknown answer in a vast data set, children seem to abruptly end their search, both when they find a reasonable (but not necessarily best) answer, and when they find none.

Of course, this does not mean that search engines need be the enemies of inquiry for children. Aside from solving the interface barriers for children (which are within the realm of near-possibility) and developing project-based pedagogy in which students work to design search engines and underlying databases to better understand the mechanics of search, only the social practice of search remains to be addressed. Moraveji speaks to this need in his proposals for the social transfer of web search expertise, which take the form of joint attention to successful search terms and URLs in co-located search activities, as well as providing access to crowd-sourced, expert search trails in distributed environments (Moraveji, 2010). Although these approaches are likely critical to a search-based pedagogy, they do not provide sufficient protection from the mechanism warned of by Dewey. There is a social component that is missing from these proposals, which is hinted at by Druin et al's discovery that children's overall perception of the Google search engine was positive. It is likely that this perception is socially learned, and was indeed for one child, who informed the researchers that his grandfather told him it was a cool site where lots of information could be found (Druin et al, 2010).

The social component of search is what provides us with the “what,” “when,” and “why” of searching, as opposed to the "how" that is emphasized in the current literature. It stands to reason that the impetus to search can only be inculcated by participating, legitimately, in a community of practice (Lave & Wenger, 1991). I experienced this personally in my formative years as a software engineer. Although I was an expert searcher in terms of “how” to search for desired data, I learned by working in close proximity to more experienced near-peers that I had a relatively weak understanding of “what,” “when,” and “why” to search in relation to my occupation. For example, learning “when” to search for existing tools – the “what” - rather than devoting development time to reinventing the wheel because the company valued cost-saving efficiency over personal innovation – the “why” - was an early, and cage-rattling lesson. Nevertheless, through this experience, my understanding of the role of search in my field of practice was heightened, as was my tendency toward inquiry-based research at work. This social mediation of the paradigms of search must also be accounted for in search-based pedagogy in order to avoid the trap of encouraging mechanistic thinking.

Dewey, J. (1916). Democracy and education, Chapter 11: Experience and thinking. New York: Macmillan (pp. 152-179 in the original).

Druin, A., Foss, E., Hatley, L., Golub, E. Guha, M.L., Fails, J., & Hutchinson, H. (2010, to appear). How children search the internet with keyword interfaces. CHI-2010.

Lave, J. & Wenger, E. (1991). Situated Learning. Cambridge: Cambridge University Pess.

Moraveji, N. (2010, April). User interface designs to support the social transfer of web search expertise. ACM – SIGIR – Doctoral Consortium, Geneva, Switzerland.