Home > Papers from 2012 > Maps for Bees? (again)

Maps for Bees? (again)

The fact that insects are proficient in using a combination of path integration and learned visual information for navigation leads naturally to questions about how these different types of information may be integrated. And of course, this line of thinking always leads to debate about whether insects possess cognitive maps of their environment.

In this new paper, Menzel et al address this debate using harmonic radar to track displaced bees. Their bees had been trained to visit two feeders and then in tests were displaced to novel sites. For analysis straight components of the flights were extracted. The directions of these straight components could all be reasonably accounted for by the bees recalling feeder specific homeward directions or by deriving a vector from the the known PI coordinates of a feeder and the ants current position. The claim is made that this latter behaviour represents a particularly sophisticated use of vectors likely to involve the use of a vector map.

It is always interesting to see the paths of displaced animals. However, before we can fully interpret them, I personally, would need two outstanding issues to be addressed. The first involves the potential use of visual route information by bees in these experiments. The authors discuss the (supposedly) barren terrain and the fact that the hive itself is not visible from more than 30m. However, they give little consideration to the potential for guidance by other panoramic cues, including their large artificial landmarks. I believe a more quantitative approach may be required to ‘prove’ what might be the size of the visual catchment areas for the feeders, the hive and the routes between the two.

The second issue involves the more traditional debate around the definition of a cognitive map. On this point the authors provide a very generous definition. Interestingly, they make reference to a recent modelling paper by Cruse and Wehner, which was written to explicitly show that robust navigational behaviour can come from a decentralised memory (I.e. no cognitive map is required). Here is a relevant quotation from the introduction:
” Recently, a neural net has been implemented in an attempt to simulate navigational strategies in ants and bees based on large-scale vector integration (Cruse and Wehner 2011). The basic components reflect the same assumptions as made in the Menzel et al. (2005) analysis namely different vector memories for outbound and inbound travels, identification of loci with respect to their home-directed vectors, and motivational change for outbound and inbound travels. The computational rules are considered as decentralised arguing against an integrated form of spatial memory as conceptualised in a cognitive map (Tolman 1948). However, multiple memory vectors that are respectively associated to landmark structures will lead to a network of retrieved and derived connections that is formally equivalent to a map-like memory structure (Menzel et al. 2005). The question is how complex such a network of connections is and how flexible it can be used in homing strategies.”

Menzel R, Lehmann K, Manz G, Fuchs J, Kobolofsky M and Greggers U (2012) Vector integration and novel shortcutting in honeybee navigation. Apidologie. DOI: 10.1007/s13592-012-0127-z

Categories: Papers from 2012
  1. No comments yet.
  1. No trackbacks yet.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s