Abstract
Plasticity in the honeybee brain has been studied using the appetitive olfactory conditioning of the proboscis extension reflex, in which a bee learns the association between an odor and a sucrose reward. In this framework, coupling behavioral measurements of proboscis extension and invasive recordings of neural activity has been difficult because proboscis movements usually introduce brain movements that affect physiological preparations. Here we took advantage of a new conditioning protocol, the aversive olfactory conditioning of the sting extension reflex, which does not generate this problem. We achieved the first simultaneous recordings of conditioned sting extension responses and calcium imaging of antennal lobe activity, thus revealing on-line processing of olfactory information during conditioning trials. Based on behavioral output we distinguished learners and non-learners and analyzed possible learning-dependent changes in antennal lobe activity. We did not find differences between glomerular responses to the CS+ and the CS− in learners. Unexpectedly, we found that during conditioning trials non-learners exhibited a progressive decrease in physiological responses to odors, irrespective of their valence. This effect could neither be attributed to a fitness problem nor to abnormal dye bleaching. We discuss the absence of learning-induced changes in the antennal lobe of learners and the decrease in calcium responses found in non-learners. Further studies will have to extend the search for functional plasticity related to aversive learning to other brain areas and to look on a broader range of temporal scales.
Highlights
A general question in the study of associative learning and memory is how stimulus-specific and outcome-related information is stored in the nervous system
For both learners and non-learners, we found no differences between the behavioral responses of the two subgroups trained respectively with 1-nonanol+ vs. 1-hexanol− or with 1-nonanol− vs. 1-hexanol+ (ANOVA for repeated measurements; learners: F1,16 = 0.28, NS; non-learners: F1,14 = 1.58, NS) so that results were pooled within each group
Our results show how difficult the search for the neural correlates of associative learning can be
Summary
A general question in the study of associative learning and memory is how stimulus-specific and outcome-related information is stored in the nervous system. Neural correlates of memory traces are difficult to delimit because changes in neural activity resulting from even simple learning forms may be distributed among different structures and regions of the brain This renders difficult the definition of which traces are important for the expression of behavior, at which time they are operative and how they relate to each other (Thompson et al, 1986; Squire, 1987). Invertebrate models are especially suited to tackle this challenge because they learn and memorize relevant information of their environments and because their nervous systems present a reduced number of neurons accessible to different recording techniques (Giurfa, 2007a; Menzel et al, 2007) Both levels of analysis can be combined as invertebrates are robust enough to facilitate parallel access to behavioral responses and neural recordings using various invasive techniques (Giurfa, 2007a)
Sign-in/Register to view highlights & summary 
Published Version (
Free)
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call