The following essay discusses two articles (Hecht, Schlaer, & Pirenne, 1942, and Cornsweet, 1970) based on a landmark experiment by Hecht, Shlaer, and Pirenne (1942), which set out to determine the minimum number of photons (quanta) required for someone to perceive a flash of light. From their findings, Hecht et al. conclude that, under certain precise conditions, the average person only needs to detect around ten quanta for their visual system to detect the flash. Furthermore, a rod cell in the retina of the eye needs only absorb one quantum of energy to initiate a response. This finding is supported by both experimental research conducted by the authors and integration of previous experimental data (Hecht et al., 1942). Cornsweet (1970) discusses the considerations involved in the experiment’s design, and explains the researchers’ findings in terms of spatial and temporal summation. The variability of the visual threshold is discussed with respect to physical and physiological variance, and the study’s significance is considered in light of current understandings of the processes underlying visual perception.
Introduction
Best services for writing your paper according to Trustpilot
Visible light is a form of electromagnetic radiation that can be detected by the human eye. At its most fundamental level, light is observed as discrete units of energy called photons or quanta. (Sportel, Bruxvoort, & Jadrich, 2009). Hecht, Shlaer, and Pirenne (1942) conducted an experiment to determine the minimum number of quanta needed to engage the visual system before a person with normal vision could perceive a flash of light. This experiment addressed a number of important factors involved in light perception, and had significant implications for our understanding of the processes that control visual perception (Cornsweet, 1970).
When Hecht et al. conducted their experiment it was common knowledge that retinal response involved such physiological factors as the sensitivity of a person’s rod and cone cells, and the extent of one’s peripheral vision. They also knew that the location and size of projected light in one’s visual field, the length of exposure to light, and the wavelength of the light, all affected one’s visual perception (Hecht et al., 1942). The researchers designed an experiment that entailed exposing subjects to short flashes of light of varying intensities and recording whether they reported seeing the flash. In designing the experimental methodology, Hecht et al. contrived to employ the optimal physical and physiological conditions that would allow for perception of light, so that a lower threshold of perception could be established experimentally.
The researchers aimed to determine the lower threshold in terms of the minimum number of quanta needed to have an effect on the visual perception system. Previous research had suggested possible answers to this research question (e.g. Chariton & Lea, 1929; Barnes & Czerny, 1932), though, as Hecht et al. point out, many of the previous experiments were marred by obvious errors in either experimental design or measurement. Additionally, whereas previous research had measured the number of quanta reaching the eye, Hecht et al. aimed specifically to determine the number of quanta engaging with the visual system at the molecular level, as this would reveal useful information about the physiological processes governing vision. This essay aims to discuss the experiment’s procedure and considerations, its key findings, and how their significance relates to current theories of visual processing.
Method
Design
On constructing the experiment, a number of factors known to influence the visual threshold needed to be controlled in order to optimize the conditions for light perception. Firstly, dark adaptation has a large impact on our ability to detect dim light in dark surroundings. Research has shown that adaptation to darkness involves a gradual increase in visual sensitivity, and that it takes approximately 40 minutes to fully adapt to see in low light (Cornsweet, 1970). In order to improve subjects’ detection of the small light flash, Hecht et al. accounted for this phenomenon by conducting the flash tests in darkness, and subjecting their participants to darkness for 40 minutes before beginning the experiments.
The second factor that the researchers considered was the location of the light flash in the subjects’ visual fields. In dim light, it is easier to see small dim objects, such as stars, when you are not directly focusing on them (Cornsweet, 1970). This is because of the varying distribution of rod cells found across the retina of the eye, which is sparser near the center of the retina, known as the fovea. These rod cells are found to be most densely situated 20 degrees out from the fovea, to the left of the left eye and vice versa (Cornsweet, 1970). For this reason, Hecht et al. designed the experiment such that a subject, while focusing on a fixation point, would be exposed to a flash-point located 20 degrees out from the center of their field of vision. Thus, they would consistently receive a flash-point on their retina 20 degrees peripheral to their fovea, where rod cells are found in highest concentration; this would enhance their ability to detect the flash of light.
A number of factors influencing the perception threshold are related to the concept of light intensity. This is because the threshold’s probability curve increases as a function of intensity. There is no true threshold of visual perception that marks when a flash becomes perceptible (Hecht et al., 1942). Indeed, increasing intensities yield gradually higher rates of perception. Cornsweet (1970, p. 14) explains that, “there is a range of intensities for which the likelihood of seeing the flash gradually and smoothly changes from low to high.” Thus, for their study, Hecht et al. chose to arbitrarily define the threshold as the intensity at which 60% of flashes are seen. They then recorded the threshold intensity for each of the participants in the experiment.
Another factor relating to light intensity is the size of the flash-point in a subject’s visual field, and consequently, the size of the image falling on their retina. Visual perception research has shown that larger spots of light have higher perception thresholds; that is, they need to be of higher intensity (convey more quanta) to be perceptible by the human eye (Cornsweet, 1970). Furthermore, this difference is observed as a threshold increase that begins with a spot covering 10 minutes of visual angle, and gradually rises as spot size increases. However, below 10 minutes, the threshold remains steady: a flash of a given number of quanta delivered within a one-minute spot or spread over 10 minutes will have the same threshold for perception by the human eye, and if detected, will appear as the same (Cornsweet,1970). Since the concepts of threshold and sensitivity are inversely related (Cornsweet, 1970), it follows that the eye is equally sensitive to both flashes-the same number of quanta are required to see them. Hence, the 10 minute section of visual angle marks the physiological limit to the human eye’s resolution.
There is a clear physical reason for this that relates to the anatomical structures and interconnections within the visual system. It is explained in terms of spatial summation, whereby the visual system amplifies the effects of individual rods within the retina. As Cornsweet (1970) describes, the optic nerve consists of many nerve fibers, each of which is connected to more than 300 rod cells via connecting neurons. Signals from any of the rods in a given area of the retina all feed into a particular nerve fiber, where their effects are summated into one signal. Incidentally, the size of each summation area relates to a 10 minute-diameter circular section of the visual angle (Cornsweet, 1970). So, for two flashes of light of equal intensity, the spread of light is inconsequential as long as the quanta delivered in the flash are absorbed by rods that reside in the same summation area (Cornsweet, 1970). Further, if the number of quanta absorbed in each such flashes is above the perception threshold, than two identical flashes will be detected by an observer (Cornsweet, 1970). With this knowledge, Hecht et al. chose to employ a circular flash-point 10 minutes in diameter, which would avoid lowering the eye’s sensitivity to the flash and raising the perception threshold.
Duration of the flash is another variable the researchers identified as affecting perception by way of intensity. Hecht et al. (1942, p. 821) explain it such that, “energy required to pass over the visual threshold involves an approximately reciprocal relationship between intensity and time of exposure.” Cornsweet (1970) describes experimental data subsequently collected, which show that flash durations shorter than 100 milliseconds have little to no impact on the perception threshold. Above this point, the threshold gradually increases. Cornsweet (1970) discusses the concept of temporal summation, and explains that, as long as the temporal distribution of quanta delivered in the flash does not extend over 100 milliseconds, the nature of the distribution in time has no effect on the perception threshold-the flash of light is consistently perceived. This indicates that the effect of an absorbed quantum on the visual system lasts for approximately 100 milliseconds, evidently so that it can summate with other quanta absorbed within that duration (Cornsweet, 1970). Cornsweet refers to a lack of current understanding of the physiological processes behind this effect, but notes the parallels between the effects of temporal and spatial summation processes within the visual system.
The final factor that Hecht et al. considered important for minimizing the perception threshold was the light’s wavelength. Previous research had shown that the minimum energy needed to see a flash varied depending on the wavelength of the light (Cornsweet, 1970). Specifically, data shows that relative sensitivity to the light is highest in the middle of the spectrum, and becomes less sensitive for wavelengths toward the upper and lower ends of the spectrum (Cornsweet, 1970). This data is referred to as the scotopic luminosity curve (Hecht & Williams, 1922). Hecht and Williams (1922) had previously measured the energy required for threshold perception of different wavelengths of light, and did so to a sufficient precision for use in calculating the absolute number of quanta needed to perceive light of different wavelengths. Their results showed that sensitivity is highest when viewing light with a wavelength of 510 nanometers (Hecht & Williams, 1922). This means that quanta with this wavelength are absorbed by the photosensitive molecules on the rod cells at a higher rate than other light. Known as the visual pigment, these molecules, by way of their atomic structure, resonate most with light that has a wavelength of around 510 nanometers, and accordingly, capture and absorb these quanta most readily (Cornsweet, 1970). To quote Cornsweet: “If each quantum in a flash is more likely to be captured, fewer quanta must be put into the eye in order that any particular number are captured.” In view of these data, Hecht et al. chose to utilize light flashes with a wavelength of 510 nanometers in their experiment, in order to maximize sensitivity to the flash and thus minimize the perception threshold.
Procedure
For the experiment, Hecht et al. constructed an apparatus that would control the precise location, duration, size, and wavelength of the light flash, with the above findings in mind. The testing was conducted in a darkened room, and each participant was subjected to darkness for 40 minutes, in order for them to adapt to dim light. For the testing to begin, the subject’s right eye was covered, and their head was held in place in the apparatus to keep the location of their left eye constantly precise. They were told to fixate on a dim red point in their field of view. When they were ready to receive a flash, they told the experimenter, who, without their knowledge, adjusted the apparatus to one of six predefined intensity levels. The subject then opened the shutter and exposed the light flash to their eye at a 20 degree angle from their point of focus. The flash lasted for 100 milliseconds, and created an image on the retina whose diameter was 10 minutes of the eye’s visual angle. The subject would then report whether they saw the flash. For each participant, this process was repeated around 50 times for each intensity level, and the frequency of detection was recorded.
Results
The results of this experiment revealed that the participants perceived flashes of light that ranged in intensity from 54 to 148 quanta (Hecht et al., 1942). These figures represent the energy reaching the cornea of the eye; since the researchers were more interested in the number of quanta absorbed by the visual pigment at the rods, they had to make three corrections to their data. The first accounted for the percentage of light reflected away from the cornea, and was estimated to be around 4% (Hecht et al., 1942). The second correction took into consideration the percentage of energy absorbed by the ocular media. The researchers obtained data from Ludvigh and McCarthy (1938), which estimated absorption by ocular media to be approximately 50% for light with a wavelength of 510 nanometers. The third correction was included to account for the percentage of light that reaches the rods but is not absorbed by the visual pigment. The light that is not absorbed does not contribute to vision; therefore, it does not form a part of the estimates for the perceptual threshold at the molecular level, and must be subtracted out.
To determine this proportion, Hecht et al. used established experimental data on the absorption of energy by visual pigment of different concentrations. For a selection of different maximal absorption levels (calculated from the concentration levels), they graphically prepared percentage-absorption wavelength spectra. These spectra showed that, at higher maximal absorption, visual pigment absorbed a wider range of wavelengths, which consistently centered on a maximum of 500 nanometers (Hecht et al., 1942). This was illustrated as a widening of the curve for higher maximal absorption levels. Equipped with this range of spectra curves, the researchers were able to compare them with the scotopic luminosity curve, the data of which they corrected for ocular media energy losses to transform them from representing incidence at the cornea to absorption at the retina (Hecht et al., 1942). Visual comparison allowed the researchers to identify which spectra curve best fit the luminosity curve, thus identifying the proportion of energy being absorbed by the visual pigment in the human eye. They found that the 10% absorption spectra curve best fit the luminosity spectrum, but conservatively chose a 20% absorption upper limit for their calculations (Hecht et al., 1942). The researchers were then able to compute an estimate of the perception threshold at the molecular level of the visual system by correcting their quanta readings for an 80% loss of energy at the retina. The range of 54-148 quanta absorbed at the cornea was calculated down to an upper limit of 5-14 quanta absorbed by the visual pigment at the rod cells.
These results were subsequently reinforced by deriving the threshold range statistically using Poisson probability distribution curves. This resulted in a threshold estimate range of 5-8 quanta, which corresponded significantly with the experimental data (Hecht et al., 1942).
Discussion
The threshold results are important when considered in light of the number of rod cells in the 10 minute area of the retina employed in the experiment. Hecht et al. estimated that a 10 minute area on the retina would contain around 500 rod cells. At the time of Cornsweet’s publication, this had been revised down to around 350. Regardless, the researchers’ original conclusions hold: The probability of a rod cell absorbing more than one quantum during a threshold flash is very low. As such, if 5-14 rod cells, located within a 10 minute area on the retina, each absorbed one quantum, this would result in a perceptible visual effect. In other words, a rod cell only needs to absorb one quantum to spur a reaction. This notion, which has been supported by subsequent research (Cornsweet, 1970), made the experiment by Hecht et al. particularly significant, because it highlighted the extent to which our visual processing has adapted to maximize visual perception of our environment.
The experiment’s findings also provided insights into the influence of biological versus physical variation on experimental data. The statistical comparisons with the Poisson probability distributions reveal that biological variability has little impact on the resultant threshold estimates because the distributions are unaffected by it (Hecht et al., 1942). Instead, it is the physical variability encompassing limits in the precision of stimulus and measurement that is governing the unaccounted variation in the data; at such small quanta numbers, “the physical nature of this variability determines the variation encountered between response and stimulus” (Hecht et al., 1942, p. 837). This observation challenged a fundamental tenet of psychological testing that suggests data variations observed in an organism’s response to a constant stimulus necessarily have biological origins (Hecht et al., 1942).
The research by Hecht et al. provided clues to the physiological processes underlying vision, and contributed to a scientific progression from cognitive-behavioral studies of visual perception, to research aimed at explaining vision at the molecular level (Cornsweet, 1970). In this way, the experiment impacted how researchers approached and designed studies of visual perception. Due to the researchers’ vigorous assessment of existing empirical data, their careful consideration of the factors influencing perception, precise experimental measurement, and their innovative methods for validating their findings, their research enabled new insights into the visual summation mechanisms that amplify our ability to perceive our environment, and had important implications for theories of the visual system’s evolution (Cornsweet, 1970).
Although the immediate goal of the experiment by Hecht and colleagues was simple-to quantify a threshold level for the visual perception of light-the necessary experimental design considerations required the researchers to think of visual processing in ways that transcended existing precepts. Their work consolidated and expanded understandings of the visual system and its underlying physiological processes, and informed theories of its evolutionary development (Cornsweet, 1970). For this reason, their experiment remains a significant milestone in visual perception research, not only because it highlighted the astounding sensitivity of the human visual system, but also because it paved the way for many important insights about the processes underlying the act of vision (Cornsweet, 1970).
