Lynch et al. Cognition enhancement in normal subjects output stage of the trisynaptic intra-hippocampal circuit(Sirvio produce useful effects in complex cortical circuits (Bouteiller et al.,1996).These observations accord with the broad idea that et al.,2011).However,the effects of increased EPSPs on network facilitated transmission at one connection will lead to a greater responses to rhythmic or complex stimulation are a critical and as number of cells transmitting to the next.Repeated across many yet unstudied issue. stages,each responding to the ampakine,this will produce a Evidence for enhanced throughput has also been obtained multiplier effect for drug action.This argument points to the con-in studies using in vivo analyses of hippocampal projections to clusion that ampakine-type drugs will exert much greater effects frontal cortex (Baumbarger et al,2001)and chronic record- in the long chains of glutamatergic neurons that constitute cor- ings from the output stage of hippocampus (Hampson et al., tical networks than in the much simpler circuits found at lower 1998b).The latter rat study showed that the number of cells levels of the neuraxis.Why the multiplier effect doesn't ultimately discharging during key steps in performing a complex task was result in abnormal discharges likely reflects the above mentioned substantially increased by systemic treatment with an ampakine inhibitory interneurons whose influence on projection neurons (Figure 6).Given that the recording site was the terminus of the also grows with increasing glutamatergic drive,as seen in input/ primary intra-hippocampal circuit,one can reasonably assume output measurements in conventional hippocampal slice exper-that the observed results reflected an augmentation of drug action iments.Since inhibition generated by arrival of a glutamater-through a polysynaptic network similar to those described for gic input is di-(and multi-)synaptic,there is brief widow in ampakines infused into hippocampal slices(Sirvio et al,1996). which network facilitation is operative.Sophisticated multi-scale But the possibility that augmented excitatory drive on ascend- (biophysics,synapses,neurons,and connectivity)computational ing biogenic amine systems,whether from the drugs or the work indicates the manner in which enhanced throughput can behavioral activity they produce,results in generalized increases DELAY INTERVAL firing rates during delay levers retracted 350- SAMPLE NONMATCH 300 LEFT 8 vehicle injection ast treatment drug injection 200 one lever extended both levers extended start treatment B 7 control ampakine o controls 471013161922252831 days ↑ FIGURE 6|Ampakine treatment increases neuronal activity in the shown;see Hampson et al.,1998b).(B)Chronic recordings from the same terminal stage of the intra-hippocampal network and this is hippocampal neuron collected during task performance prior to and after the accompanied by supra-normal performance on a complex task.(A)Adult start of drug treatment.As shown,the ampakine dramatically increased the rats were extensively trained in the delayed non-match to sample (DNMS) firing rate most particularly after the initial response to the sample (blue task,illustrated here,wherein they were required to sequentially (i)press a arrow)and in the delay phase leading up to the non-match choice (red arrow); bar to receive a reward (sample phase;left)at which time a light appeared on this pattern and ampakine-related change was typical of field CA1 cells.(C) the opposite wall of the arena,(ii)move to the wall with the light during the Plot shows the mean firing rate of all units recorded during the delay phase "delay phase"(top)and perform a nose poke for random intervals until the expressed as a percent of the baseline firing rate (prior to ampakine light extinguished,and then (iii)return to the original side of the arena and treatment):the progressive increase in firing rate in the experimental rats press the bar that had not been originally selected (non-match phase;right). paralleled the increase in successful trials in the same rats.Rats given vehicle The level of success in executing the task reached asymptote after weeks of only (open circles)exhibited no change in cell firing or successful task training,after which the animals were separated into vehicle or drug treated execution over the full period of testing.Note that the increased firing in the groups.Injections were given every other day.Animals given the ampakine drug group was still present on vehicle days (red dots),and indeed for the prior to individual training sessions showed a progressive increase in correct week following cessation of treatment,an effect that is interpreted as being responses over the next 2+weeks of daily testing.ultimately reaching levels due to facilitation of LTP-type plasticity on drug days.Adapted from Hampson far higher than the asymptotic level maintained by vehicle-only rats (not et al.(1998b). Frontiers in Systems Neuroscience www.frontiersin.org May 2014 Volume 8 Article 90 11Lynch et al. Cognition enhancement in normal subjects output stage of the trisynaptic intra-hippocampal circuit (Sirvio et al., 1996). These observations accord with the broad idea that facilitated transmission at one connection will lead to a greater number of cells transmitting to the next. Repeated across many stages, each responding to the ampakine, this will produce a multiplier effect for drug action. This argument points to the con￾clusion that ampakine-type drugs will exert much greater effects in the long chains of glutamatergic neurons that constitute cor￾tical networks than in the much simpler circuits found at lower levels of the neuraxis. Why the multiplier effect doesn’t ultimately result in abnormal discharges likely reflects the above mentioned inhibitory interneurons whose influence on projection neurons also grows with increasing glutamatergic drive, as seen in input / output measurements in conventional hippocampal slice exper￾iments. Since inhibition generated by arrival of a glutamater￾gic input is di- (and multi-) synaptic, there is brief widow in which network facilitation is operative. Sophisticated multi-scale (biophysics, synapses, neurons, and connectivity) computational work indicates the manner in which enhanced throughput can produce useful effects in complex cortical circuits (Bouteiller et al., 2011). However, the effects of increased EPSPs on network responses to rhythmic or complex stimulation are a critical and as yet unstudied issue. Evidence for enhanced throughput has also been obtained in studies using in vivo analyses of hippocampal projections to frontal cortex (Baumbarger et al., 2001) and chronic record￾ings from the output stage of hippocampus (Hampson et al., 1998b). The latter rat study showed that the number of cells discharging during key steps in performing a complex task was substantially increased by systemic treatment with an ampakine (Figure 6). Given that the recording site was the terminus of the primary intra-hippocampal circuit, one can reasonably assume that the observed results reflected an augmentation of drug action through a polysynaptic network similar to those described for ampakines infused into hippocampal slices (Sirvio et al., 1996). But the possibility that augmented excitatory drive on ascend￾ing biogenic amine systems, whether from the drugs or the behavioral activity they produce, results in generalized increases FIGURE 6 | Ampakine treatment increases neuronal activity in the terminal stage of the intra-hippocampal network and this is accompanied by supra-normal performance on a complex task. (A) Adult rats were extensively trained in the delayed non-match to sample (DNMS) task, illustrated here, wherein they were required to sequentially (i) press a bar to receive a reward (sample phase; left) at which time a light appeared on the opposite wall of the arena, (ii) move to the wall with the light during the “delay phase” (top) and perform a nose poke for random intervals until the light extinguished, and then (iii) return to the original side of the arena and press the bar that had not been originally selected (non-match phase; right). The level of success in executing the task reached asymptote after weeks of training, after which the animals were separated into vehicle or drug treated groups. Injections were given every other day. Animals given the ampakine prior to individual training sessions showed a progressive increase in correct responses over the next 2+ weeks of daily testing, ultimately reaching levels far higher than the asymptotic level maintained by vehicle-only rats (not shown; see Hampson et al., 1998b). (B) Chronic recordings from the same hippocampal neuron collected during task performance prior to and after the start of drug treatment. As shown, the ampakine dramatically increased the firing rate most particularly after the initial response to the sample (blue arrow) and in the delay phase leading up to the non-match choice (red arrow); this pattern and ampakine-related change was typical of field CA1 cells. (C) Plot shows the mean firing rate of all units recorded during the delay phase expressed as a percent of the baseline firing rate (prior to ampakine treatment): the progressive increase in firing rate in the experimental rats paralleled the increase in successful trials in the same rats. Rats given vehicle only (open circles) exhibited no change in cell firing or successful task execution over the full period of testing. Note that the increased firing in the drug group was still present on vehicle days (red dots), and indeed for the week following cessation of treatment, an effect that is interpreted as being due to facilitation of LTP-type plasticity on drug days. Adapted from Hampson et al. (1998b). Frontiers in Systems Neuroscience www.frontiersin.org May 2014 | Volume 8 | Article 90 | 11