New neurons are continuously generated in the adult mammalian olfactory bulb. The role of these newborn neurons in olfactory learning has been debated. Blocking the addition of neurons has been reported either to result in memory alteration or to have no effect at all (Imayoshi et al., 2008; Breton-Provencher et al., 2009; Lazarini et al., 2009; Sultan et al., 2010). These discrepancies may have arisen from differences in the behavioral paradigms used: operant procedures indicated that neurogenesis blockade had substantial effects on long-term memory (Lazarini et al., 2009; Sultan et al., 2010) whereas other methods had little effect (Imayoshi et al., 2008; Breton-Provencher et al., 2009). Surprisingly, while operant learning is known to modulate the survival of new neurons, the effect of non-operant learning on newborn cells is unknown. Here we use mice to show that compared with operant learning, non-operant learning does not affect cell survival, perhaps explaining the current controversy. In addition, we provide evidence that distinct neural substrates at least partly underlie these two forms of learning. We conclude that the involvement of newborn neurons in learning is subtly dependent on the nature of the behavioral task.