A temporal hierarchical feedforward model explains both the time and the accuracy of object recognition

Abstract

Brain can recognize different objects as ones it has previously experienced. The recognition accuracy and its processing time depend on different stimulus properties such as the viewing conditions, the noise levels, etc. Recognition accuracy can be explained well by different models. However, most models paid no attention to the processing time, and the ones which do, are not biologically plausible. By modifying a hierarchical spiking neural network (spiking HMAX), the input stimulus is represented temporally within the spike trains. Then, by coupling the modified spiking HMAX model, with an accumulation-to-bound decision-making model, the generated spikes are accumulated over time. The input category is determined as soon as the firing rates of accumulators reaches a threshold (decision bound). The proposed object recognition model accounts for both recognition time and accuracy. Results show that not only does the model follow human accuracy in a psychophysical task better than the well-known non-temporal models, but also it predicts human response time in each choice. Results provide enough evidence that the temporal representation of features is informative, since it can improve the accuracy of a biologically plausible decision maker over time. In addition, the decision bound is able to adjust the speed-accuracy trade-off in different object recognition tasks.

Linear Integration of Sensory Evidence over Space and Time Underlies Face Categorization

“…Past research on object recognition focused largely on feedforward visual processing and instantaneous readout of the visual representations, leaving a conceptual gap for understanding the temporally extended processes that underlie perception and action planning based on visual object information. Several studies have attempted to fill this gap by using noisy object stimuli (Heekeren et al, 2004;Philiastides and Sajda, 2006;Ploran et al, 2007;Philiastides et al, 2014;Heidari-Gorji et al, 2021) or sequential presentation of object features (Ploran et al, 2007;Jack et al, 2014). However, the stimulus manipulations in these studies did not allow a comprehensive exploration of both spatial and temporal processes.…”

A fully spiking coupled model of a deep neural network and a recurrent attractor explains dynamics of decision making in an object recognition task

“…• Since these regression models were widely used to explain human and animal behaviors in decision science [47], their success in mimicking the behavior of the proposed model can support the plausibility of the proposed model. • It also shows that similar to human [24] and animal behaviors during perceptual decision making task [28], reaction time (equation ( 8 For example, for the threshold is 15 and the strongest stimulus (red line, stimulus strength = 100%), the model can perfectly distinguish whether the stimulus is face or car. On the other hand, in the case of the weakest stimulus, the observation time of the input image for the model, is longer, resulting in the using of more levels of information for the decision making.…”

“…This has been done by resetting each neuron after generating a spike and also permitting the neuron to generate multiple spikes. The current study is different from [24] in terms of using spiking deep neural networks which has been shown to be more biologically plausible than spiking HMAX [24] in a lot of aspects.…”

“…In this study, in line with our recent works [23,24], we present a time-based multilevel model for object recognition using a spiking deep neural network coupled with a spiking recurrent attractor based decision making model. In the first stage of the model (temporal feeding), the input image information is sent to the next layer in a temporal representation.…”

“…To account for the speed-accuracy trade-off regime in object recognition we used a well-known attractor-based decision-making neural model and coupled it with the proposed temporal spiking deep neural network. Although some recent models (as our recent work [23,24]) used the accumulation to bound mechanism [20,[22][23][24] to account for speed-accuracy trade-off neither of them investigate if the neural implementation of accumulation to bound model is able to explain dynamics of object recognition in the brain. This is important because the neural implementation of accumulation to bound model (or attractor based recurrent neural networks) cannot exactly behave as the accumulation to bound process [63,64].…”