Keywords

1 Introduction

The “probe” technique [1, 2] aims to support user-centred design by facilitating the exploration and collection of new ideas concerning technology use and services with a participatory design orientation. This technique consists essentially in introducing and providing targeted “potential” users with one (or more) simple artefact(s) illustrative of the considered emerging technology so that the participants can put and use it at home, in their daily environment for a significant period of time, i.e. several days or even weeks. The main assumption is that users’ ideation is stimulated by handling the artefact and projecting possible uses from their individual’s daily physical and social environment. Moreover, this technique facilitates the collection of evidence by the participants regarding their contexts and the potential uses associated.

2 Study Goal and Design

This study examines to what extend the «probe» technique can support users in their elicitation of creative uses as possible precursors of «latent needs» related to an emerging technology. The technique is compared with a standard prototype exploration situation where subjects came to the laboratory to be introduced to, to manipulate and to test the artefact during 15 min before eliciting creative uses. The artefact that act as the “technical probe” was a plug and play pico-projector (iptek T30; cf. Fig 1) compatible with Iphones and any PC having a S-Video, that could be used either with a tripod or hand held. It was chosen as representative of emerging technologies because it was not well known by the public at the moment of the study (spring 2013). Two eliciting conditions (individual vs. group) were contrasted in order to investigate the effect of social construction on the generation of creative ideas.

Fig. 1.
figure 1

Pictures of the pico projector used as the technical probe in this study

Full size image

3 Method

3.1 Participants

140 participants (120 females and 120 males) regular users that own a smartphone were recruited through advertisement in various specialised fora (e.g iphonesoft.fr) and social networks (Facebook et Twitter). All participants were living in Paris area. They were not paid but participated in a draw to win one amongst 20 pico-projectors.

3.2 Procedure

In a first phase, participants in the “probe” condition came to pick up the pico-projector at the laboratory premises. Before going back home with the artefact, they were shown and briefly explained how it works. After one week, they had to bring it back to the university laboratory for the next phase. Participants in the “prototype test” condition came to the laboratory premises and were similarly shown and explained the pico-projector. Then, they were invited to use it freely and explore its functions and possible uses for a quarter of an hour.

In the second phase, the participants were asked to write down on dedicated paper sheets as many ideas of future services involving the pico-projector as they could and how these services could respond to some of their needs. It was made clear they were not expected to refer only to the uses they optionally did during the previous phase of the study. In the individual production condition, participants were instructed to imagine future projection-based services with the mini-projector and imagine how that could meet their needs. In the collective condition, participants were grouped by four. They received the same instruction as in the individual condition, but the fact that they were encouraged to discuss and share the idea.

Whatever the condition, we recalled the general principles for the production phase as follows: proposed solutions should not be criticized, generate as many solutions as possible (quantity is better than quality), all ideas count especially the most original, one may enhance some of previously generated solutions to generate new ones.

3.3 Collected Data

In total, we collected 1525 ideas. Each one was written on a paper sheet with structured fields (see Fig. 2).

Fig. 2.
figure 2

Two examples of collected structured reports for ideas in the production phase

Full size image

3.4 Data Analysis

We analysed the effect of conditions both in terms of fluency and quality of generated ideas, i.e. degree of creativeness. The latter was measured by 12 judges using the following procedure. In the first step, the set of 1525 ideas was randomly divided into 4 subsets (3 subsets of 381 ideas and 1 of 382), each being rated by a group of 3 judges on a scale from 1 (not creative at all) to 7 (extremely creative). The inter-rater agreement was verified using the two-way mixed Intra-Class Coefficient (ICC) for single measurements (Table 1) and showed a high degree of agreement within each group of judges.

Table 1. Intra-Class Coefficient (ICC) for the four groups of experts
Full size table

In the second step, we extract the thirty most creative ideas from each subset resulting in a final set of 120 ideas. These ideas were further evaluated by the same judges using the subscales of 3 dimensions of the Creative Product Semantic Scale [3]: Novelty, Resolution, and Elaboration and Synthesis. The analysis demonstrated a high level of internal consistency for each of three dimensions whatever the group (Table 2). Furthermore, we found a high inter-rater agreement between groups for group 1 (r = 0.87, p < 0.05), group 3 (r = 0.96, p < 0.05) and group 4 (r = 0.81, p < 0.05). Oppositely, group 2 exhibited a poor agreement (r = 0.14, p > 0.05).

Table 2. Internal consistency of creativity dimensions for the four groups (Cronbach’s coeff.)
Full size table

4 Results

4.1 Number of Produced Ideas

Fluidity was evaluated by counting the number of distinct ideas produced by participants. Descriptively, we observe that the Probe condition results in more ideas than the laboratory condition, and that the individual production also results in more ideas (Table 3) than the collective one. The expositions and productions factors show not significant correlations (χ2 (1) = 2.9313, p = .085).

Table 3. Distribution of the 1525 idea across the four experimental conditions
Full size table

Table 4 reports on the average number of ideas produced by participants depending the exposition and generation conditions. Participants elaborated significantly more ideas in the “probe” condition bringing the device at home for one week (mNprobe = 4.11, SD = 2.26) than in the “laboratory prototype test” condition (mNlab = 3.75, SD=2.20 F(1,1521) = 15.56, p < .00). They also elaborated significantly more ideas in the individual condition (mNindiv = 4.36, SD = 2.41) that in the group condition (mNgroup = 3.37, SD = 1.82; F(1,1521) = 83.41, p < 0.001). The two factors exhibit no significant interaction (F(1,1521) = 3.77, p < . 052 n.s.).

Table 4. Average number of ideas produced according to the initial condition of exposition, to the technology (Probe vs. Laboratory) and condition of production (Individual vs. collective)
Full size table

4.2 Creativity of the Generated Idea

In terms of the quality of generated idea, the ideas generated in the Probe conditions were significantly rated as more creative (mProbe = 4.21, SD = 1.35) than those generated in the laboratory condition (mLab = 3.63, SD = 1.36; F(1, 1521) = 135.21, p < .000). There was no significant main effect of the production condition (mindiv = 3.92, SD = 1.42; mQgroup = 3.97, SD = 1.32; F(1,1521) = 0.02, p < .967 n.s.) but a significant two ways interaction between the two factors (F(1,1521) = 18.42, p < .001). This reveals that for “probe” participants, the level of idea creativity is higher in the group production condition (mprobeGroup = 4.33, SD = 1.32) than in the individual production condition (mprobeIndiv = 4.11, SD = 1.36) whereas the opposite is observed for “laboratory prototype test” group. The individual production group exhibited more creative idea (mLabIndiv = 3.50, SD = 1.16) than in the collective condition (mLabGroup = 3.72, SD = 1.47) (Table 5).

Table 5. Average level of creativity of ideas produced according to the condition of exposition to the technology (Probe vs. Laboratory) and idea production (Individual vs. Collective)
Full size table

As a complementary step, we looked at the distribution of the thirty highest creative solutions across the conditions by our four group of experts (Table 6). We observed that a significant higher number of these creative ideas were found in the Probe x Collective group (χ2 (3) = 25.8667, p < .000). Furthermore, the 4 experts groups did not significantly differ in how they distribute creative vs. less creative idea across the four conditions (data not reported; χ2 (9) = 3.34, p = .095).

Table 6. Distribution of 30 most creative identified by the 4 group of experts across the four conditions
Full size table

5 Discussion and Conclusions

Users generate more creative uses and needs in free and active exploration of introduced artefact issued from emerging technologies in their own daily environment than in laboratory test condition. As shown in this first study, the “probe” method has is effective to support users’ creativity and to help them imagining service and new needs that emerging technologies could fulfil. Actually, we found that it exceeds the in-laboratory prototype-based production condition both in terms of quantity and quality of ideas. This results may be explained by several factors that could be further investigated in the future, like: at home, the artefact can be put and used in numerous and diverse contexts that provide the participant with multiple affordances and stimulate the production of wider range of new uses; participants in the probe condition can use the artefact with members of their family and friends, which increases the potential of mixing viewpoints and individual contributions while testing and co-constructing ideas; finally, it is possible that the laboratory situation in presence of the experimenter and under time constraints can be less fun and motivating than being at home as in the probe condition. In the “probe” group, there is potentially a longer time of exposition to the artefact and of possible use and thinking about new uses, although participants may as well have simply left the artefact in a drawer somewhere during the one-week period.

We also replicated the result widely reported in the literature that individuals working alone produced significantly more idea than those working interactively within a group [4; see e.g. the review by 5], which can be caused by an effect of social loafing and free riding [6] as well as of cognitive overload [7].