Intuitively, one may expect that the reaction to novelty is a general property of an individual applicable across a wide range of situations – an individual that readily investigates a change in the environment, such as a newly emerging food source, may also hesitate little to investigate an unfamiliar environment beyond its territory. However, reactions to novelty are often context specific. For example, individual Pumpkinseed-sunfish (Lepomis gibbosus) that more rapidly approached a meter-stick were not necessarily the first to investigate a novel food source (Coleman & Wilson Reference Coleman and Wilson1998). In Carib grackles (Quiscalus lugubris), levels of neophilia (attraction to novelty) were correlated within different contexts (investigating unfamiliar objects, tasting novel food, or entering unfamiliar space), but not across contexts (Reader Reference Reader2003). In genera as different as parrots and songbirds, neophilia differs across closely related species in a context-dependent manner depending on the ecology and migratory strategy of a species. For example, when compared to migratory species, year-round resident species explore changes in their familiar environment more rapidly and thoroughly (Mettke-Hofmann Reference Mettke-Hofmann2007; Mettke-Hofmann et al. Reference Mettke-Hofmann, Ebert, Schmidt, Steiger and Stieb2005a; Reference Mettke-Hofmann, Wink, Winkler and Leisler2005b), whereas migratory rather than resident species are more likely to explore a novel room (Mettke-Hofmann et al., under review). These differences in neophilia have likely evolved as an adaptation to the migratory strategy of the species (Mettke-Hofmann et al. Reference Mettke-Hofmann, Wink, Winkler and Leisler2005b). Likewise, still-invading populations of house sparrows (Passer domesticus) feed from a novel food source more quickly than long-established populations, even though the two populations showed similar levels of neophobia (avoidance of novelty) when a novel object was presented next to familiar food (Martin & Fitzgerald Reference Martin and Fitzgerald2005).

Given that novelty reactions (a) differ intraspecifically across contexts, (b) are interspecifically differently expressed across contexts depending on the ecology and migratory strategy of a species, and (c) are the main underlying mechanisms of innovations, it is likely that innovations emerge predominantly in contexts in which an individual or species shows the most neophilia. This raises the following question. Are neophilic reactions in all contexts equally likely to favor innovations, or are neophilic reactions in one context more likely to lead to innovations than those in other contexts (contextual bias)? Findings from Reader (Reference Reader2003) in the Carib grackle seem to favor the latter. The number of different attempts to solve a foraging task (a measure of innovation) was most closely associated with object neophilia (approaching an unfamiliar object) than with neophilia to taste a novel food or enter a novel room. However, the foraging task involved manipulating a Plexiglas box to reach food (Webster & Lefebvre Reference Webster and Lefebvre2001). What if the foraging task had involved a spatial context, such as finding a way into a familiar room with food (e.g., trying several entrances)? Would there have been an association with spatial neophilia?

The fact that many of the innovations described in the literature include manipulating items, tool use, or novel food suggests that object and food neophilia may be more prone to induce innovations than spatial neophilia. This may also reflect a bias for detecting innovations towards the former context, however. Object- or food-related innovations are probably easier to identify than spatial innovations such as use of a different strata or movement technique, or the use of a novel habitat. In birds, only 13% of the reported innovations (augmented version of the database used in Lefebvre et al. Reference Lefebvre, Reader and Sol2004) can be classified as spatial innovations, although birds may be more likely than other taxa to show spatial innovations because of their three-dimensional use of the environment. In Bornean Orangutans (Pongo pygmaeus wurmbii; van Schaik et al. Reference van Schaik, van Noordwijk and Wich2006), 5% of the 43 innovations reported classify as spatial innovations.

To solve the problem of contextual versus detection bias we need studies that (1) pay special attention to spatial innovations and (2) investigate the relationship between neophilia and innovations within and between contexts across a broad range of species. Resident-migrant comparisons are ideally suited to address this problem as residents show more object neophilia and migrants more spatial neophilia (Mettke-Hofmann Reference Mettke-Hofmann2007; Mettke-Hofmann et al. Reference Mettke-Hofmann, Ebert, Schmidt, Steiger and Stieb2005a; Reference Mettke-Hofmann, Wink, Winkler and Leisler2005b). Resident palaearctic species are also more innovative than migratory counterparts (Sol et al. Reference Sol, Lefebvre and Domingo Rodriguez-Teijeiro2005b). However, it is currently unknown whether residents have a higher innovation rate than migrants due to a contextual bias for innovations to emerge preferentially in object-related contexts, or whether the former just have more object-related innovations, whereas the latter may innovate equally often or even more often in spatial contexts, although we miss it because of a detection bias.

In addition to object-related versus spatial innovations, such comparisons may also include object-related versus food-related innovations. Distinguishing between object-related and food-related innovations is important, as shown in the study of the Carib grackle (Reader Reference Reader2003) where object neophilia but not food neophilia was mostly associated with the foraging task. Similarly, differences in food neophobia but not object neophobia distinguished expanding and established house sparrow populations (Martin & Fitzgerald Reference Martin and Fitzgerald2005).

Investigating the relationship between novelty reactions and innovations contributes to our understanding of the underlying mechanisms of innovations and allows us to predict the context in which innovations are most likely to occur. To accurately compare innovativeness across species, as suggested by Ramsey et al., we not only have to take into consideration differences in social transmission, but also the fact that biases may exist towards detecting innovations more frequently in certain contexts. Awareness of the context-specifity of innovations further helps to make comparisons more reliable. Consequently, interspecific comparisons should control for the context in which innovations occur by restricting comparisons to object-related or food-related innovations until other forms of context-dependent innovation have been systematically measured.