Using electroretinograms to assess flicker fusion frequency in domestic hens Gallus gallus domesticus

Vision Research 62 (2012) 125–133

Contents lists available at SciVerse ScienceDirect

Vision Research journal homepage: www.elsevier.com/locate/visres

Using electroretinograms to assess flicker fusion frequency in domestic hens Gallus gallus domesticus Thomas J. Lisney a,⇑, Björn Ekesten b, Ragnar Tauson c, Olle Håstad d,1, Anders Ödeen a a

Department of Animal Ecology, Evolutionary Biology Centre, Uppsala University, Norbyvägen 18D, SE-752 36 Uppsala, Sweden Department of Clinical Sciences, Swedish University of Agricultural Sciences, P.O. Box 7054, SE-750 07 Uppsala, Sweden c Department of Animal Nutrition and Management, Swedish University of Agricultural Sciences, Kungsängen Research Centre, SE-753 23 Uppsala, Sweden d Department of Evolutionary Organismal Biology, Uppsala University, Norbyvägen 18A, SE-752 36 Uppsala, Sweden b

a r t i c l e

i n f o

Article history: Received 26 July 2011 Received in revised form 16 March 2012 Available online 11 April 2012 Keywords: Animal welfare Bird Chicken Critical flicker fusion frequency Temporal resolution

a b s t r a c t The assessment of flicker fusion frequency (FFF), the stimulus frequency at which a flickering light stimulus can no longer be resolved and appears continuous, and critical flicker fusion frequency (CFF; the highest frequency at any light intensity that an observer can resolve flicker) are useful methods for comparing temporal resolution capabilities between animals. Behavioural experiments have found that average CFFs in domestic chickens (Gallus gallus domesticus) are in the range of ca. 75–87 Hz, measured in response to full spectrum (i.e. white light plus UV) stimuli. In order to examine whether the chicken retina is able to detect flicker at higher frequencies, we used electroretinograms (ERGs) to assess FFF/CFF in adult hens from two commercial genotypes, Lohmann Selected Leghorns (LSLs) and Lohmann Browns (LBs). ERGs were recorded in response to flickering light at ten full spectrum light intensities ranging from 0.7 to 2740 cd m 2. Two methods were used to determine FFF/CFF from the ERG recordings and these methods yielded very similar results, with average FFF ranging from ca. 20 Hz at 0.7 cd m 2 to an average CFF of ca. 105 Hz at 2740 cd m 2. In some individuals, CFFs of 118–119 Hz were recorded. The Intensity/FFF (I/FFF) curves are double-branched with a break point representing the rod-cone transition occurring between 2.5 and 5.9 cd m 2. No significant differences in the I/FFF curves were found between the two genotypes. At stimulus light intensities >250 cd m 2, the ERG-derived FFF and CFF values are all higher than those from behavioural studies using the same stimuli. Although hens do not appear to be able to consciously perceive flicker above approximately 90 Hz, the finding that the ERG responses are able to remain in phase with light flickering at frequencies >100 Hz means that the retinae of domestic poultry housed in artificial light conditions may be able to resolve flicker from fluorescent lamps. As range of detrimental effects have been reported in humans as a result of exposure to such ‘‘invisible flicker’’, the possibility exists that flicker from fluorescent lamps also acts as stressor in domesticated birds. Ó 2012 Elsevier Ltd. All rights reserved.

1. Introduction A fundamental characteristic of a visual system is its ability to resolve temporally varying stimuli. Temporal resolution can be determined by measuring the flicker fusion frequency (FFF), which is the stimulus frequency at which a flickering light stimulus can no longer be resolved and appears continuous. The critical flicker fusion frequency (CFF) is the highest FFF, irrespective of light

⇑ Corresponding author. Present address: Department of Psychology, University of Alberta, Edmonton, Alberta, Canada T6G 2E9. Fax: +1 780 492 1768. E-mail addresses: [email protected] (T.J. Lisney), [email protected] (B. Ekesten), [email protected] (R. Tauson), [email protected] (O. Håstad), [email protected] (A. Ödeen). 1 Present address: Department of Anatomy, Physiology and Biochemistry, Swedish University of Agricultural Sciences, Uppsala, Sweden. 0042-6989/$ - see front matter Ó 2012 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.visres.2012.04.002

intensity, and this measure is often used to compare temporal resolution in vertebrates. Among vertebrates, birds are suspected to have high CFFs, because most species are highly active during photopic (daylight) conditions and need to be able to detect and process fast moving stimuli (Greenwood et al., 2004; Jones, Pierce, & Ward, 2007; Meyer, 1977). While there is currently little experimental evidence to support this contention (Lisney et al., 2011), it should be noted that the highest CFF reported for a vertebrate (143 Hz) was recorded in the pigeon Columba livia (Dodt & Wirth, 1953; Meyer, 1977). In recent years a number of researchers have investigated FFF/ CFF in the domestic chicken Gallus gallus domesticus (Jarvis et al., 2002; Lisney et al., 2011; Railton, Foster, & Temple, 2009; Rubene et al., 2010). This is because the chicken is commonly used as a model to research human eye disease (Smith, 1992) and because of concerns over welfare in domestic poultry, where birds are

126

T.J. Lisney et al. / Vision Research 62 (2012) 125–133

housed in artificial light conditions created by fluorescent lamps (Nuboer, Coemans, & Vos, 1992). Such lamps flicker at twice the rate of the alternating current electricity supply, resulting in a flicker rate of 100 Hz in Europe or 120 Hz in North America (Wilkins & Clark, 1990) that may be perceived by the avian visual system and so act as a stressor (Greenwood et al., 2004; Maddocks, Goldsmith, & Cuthill, 2001; Nuboer, Coemans, & Vos, 1992; Prescott, Wathes, & Jarvis, 2003; Widowski, Keeling, & Duncan, 1992). FFF/CFF in animals can be determined using electrophysiological techniques such as the electroretinogram (ERG), which records electrical potentials generated in the retina, or behavioural methods (Brown, 1965; D’Eath, 1998; Douglas & Hawryshyn, 1990). Although there is often congruence, the results of ERG and behavioural studies can yield differences in flicker sensitivity. This is because the ERG reflects the neural activity of the retina, while FFF/ CFF values obtained from behavioural testing reflect what an animal actually perceives and are a result of visual processing further along the visual pathway in the brain (Douglas & Hawryshyn, 1990; Schneider, 1968). In chickens, a number of recent studies have used behavioural methods to assess FFF/CFF (Jarvis et al., 2002; Lisney et al., 2011; Prescott, Wathes, & Jarvis, 2003; Railton, Foster, & Temple, 2009; Rubene et al., 2010). These studies have reported average CFFs, measured in response to white, broad-spectrum light stimuli, in the range of ca. 75–87 Hz. The aim of this research was to complement this previous body of work by using the ERG to assess flicker sensitivities in domestic chickens across a wide range of light intensities (I), in order to create an ‘I/FFF curve’ (Henkes, 1964; Lisney et al., 2011). The ERG provides an indication of the maximum possible flicker detection rate of the eye at the level of the retina, prior to temporal summation that may occur further along the visual pathway (D’Eath, 1998). Also, as most of the I/FFF curves and CFF values reported in the literature for various animals, including birds, were recorded using ERGs (e.g. Dodt & Enroth, 1954; Dodt & Wirth, 1953; Ordy & Samorajski, 1968) the data presented here should allow for better ‘like-for-like’ comparisons between the domestic chicken and other species, rather than I/FFF curves derived from behaviour. 2. Materials and methods 2.1. Animals Four non-beak trimmed laying hens of each of the two commercial genotypes LSL (Lohmann Selected Leghorn) and LB (Lohmann Brown) (eight birds in total) were used in this study (Table 1). The hens were aged 69–73 weeks and weighed between 1.63 and 2.12 kg at the time of the experiments. The hens came from a layer unit were they were housed in groups of eight in Victorsson furnished cages (Bröderna Victorsson AB, Frillesås, Sweden) installed in three tiers. Each cage featured a nest, perches and a litter box

(described in further detail by Tauson and Holm (2001)). The hens were given ad lib a commercial layers’ feed in the form of crumbles from an automatic flat chain feeder in the feed trough three times daily. Litter provided as saw dust was replenished at least once a week. Manure was collected on endless belts under each tier of cages and removed twice a week. Eggs were collected manually daily. Water was provided from nipple racks at the rear of each cage. Incandescent light was provided from 30% dimmed white 60 W bulbs enclosed in Hessling lamps hanging between the rows of cages. The light period was increased from 9 h per day (07.00– 16.00) at 15 weeks of age until 14 h (03.00–17.00) at 23 weeks of age and onwards. Luminance was increased from ca. 0–7 lx over a 30 min time course at lights-on in the morning to imitate dawn, and dimmed over a similar time course in the evening to imitate dusk. The light intensities were purposely kept low to reduce feather pecking (Kjaer & Vestergaard, 1999). 2.2. Anaesthesia Each hen was anaesthetised with an intraperitoneal (IP) injection of 37 mg kg 1 ketamine hydrochloride (Ketaminol vet., Intervet AB, Sollentuna, Sweden) and 6 mg kg 1 xylazine (Narcoxyl vet., Intervet AB, Sollentuna, Sweden) (Wortel, Rugenbrink, & Nuboer, 1987) and anaesthesia was maintained via supplementary injections of smaller doses the same drugs in the same proportions, delivered approximately every 20–30 min after the initial dose, via an IP catheter. Each hen was placed inside a Faraday cage, and secured in a padded, adjustable holder. This allowed the head to be positioned so that the cornea of the right eye was held parallel to and directly under (at a distance of 12 mm) the stimulus delivery system. Reusable hand-warmers were used to maintain body temperature. The eyelids were held apart using a lid speculum and oxibuprocaine hydrochloride eye drops were used for additional, topical anaesthesia of the cornea (Oxybuprocaine Chauvin, Novartis Pharma AG, Basel, Switzerland). Artificial tears containing hyaluronic acid (ZilkEye, Evolan Pharma AB, Danderyd, Sweden) were applied regularly during each experiment to keep the cornea moist. After each experiment the hens were euthanised with an overdose of sodium pentobarbital. The experiments were performed between 12:00 and 18:00 (Swedish local time: CET) as the chicken ERG is known to show circadian effects (Lu, Zoran, & Cassone, 1995). All of the experiments were pre-approved by the regional ethical committee (Uppsala djurförsöksetiska nämnd) and adhered to the ARVO guidelines for animal experimentation. 2.3. Stimulus The stimulus delivery system was adapted from that described by Lisney et al. (2011) and is illustrated in Fig. 1. Six light-emitting diodes (LEDs) (5 mm, 30 deg) were arranged at the distal end of a 40 mm long, 20 mm diameter aluminium tube. White (Avago technologies, Malaysia) and UV (single peak at 400 nm, Hero, South

Table 1 Age and body mass information for the eight hens used in this study and the order of stimulus presentation for each hen. LB and LSL refer to the genotypes Lohmann Brown and Lohmann Selected Leghorn, respectively. Experiment order

Hen

Age (weeks)

Body mass (kg)

Order of stimulus presentation

1 2 3 4 5 6 7 8

LB1 LSL1 LSL2 LSL3 LB2 LSL4 LB3 LB4

69 70 70 71 71 72 72 73

1.96 1.63 1.87 1.78 2.12 1.86 2.00 1.90

Group Group Group Group Group Group Group Group

1, 1, 2, 1, 2, 2, 1, 2,

Group Group Group Group Group Group Group Group

2 2 1 2 1 1 2 1

127

T.J. Lisney et al. / Vision Research 62 (2012) 125–133

8 1 2 7 3

4

6 5 Fig. 1. Schematic diagram of the stimulus delivery system used in this study. Six LEDs (1) were arranged at the distal end of a 40 mm long, 20 mm diameter aluminium tube (2). This aluminium tube was inserted into a dark grey matte plastic cylinder (3). Different combinations of neutral density filters could be attached to the proximal end of this cylinder (4). This plastic cylinder in turn was placed into a second, wider bore housing cylinder (5) made from the same plastic, into which a UV-transparent Perspex panel in combination with a diffusion filter was inserted (6). The stimulus delivery system was held in position above the hen using a clamp stand (7). The LEDs were connected to a function generator (8).

Korea) LEDs were combined in a 2:1 ratio in order to create a ‘full spectrum’ stimulus as described in detail elsewhere (Lisney et al., 2011; Rubene et al., 2010). The LEDs were connected to a function generator (2 MHz, GFG-8020H, GW Instek, Suzhou, China), which was used to create square wave, 100% modulation flickering stimuli (with a 50% duty cycle) of varying frequency. The aluminium tube containing the LEDs was inserted into a dark grey matte plastic cylindrical housing. Different combinations of neutral density (ND) filters (Lee Filters, Andover, UK; see below) could be attached to the proximal end of this cylinder, which was in turn was placed into a second, wider bore housing cylinder made from the same plastic, which was held in position above the hen using a clamp stand. The light stimuli created by the LEDs were projected onto the right eye through a UV-transparent Perspex panel in combination with a diffusion filter (Lee Filters) positioned at the proximal end of the outer cylindrical housing held in the clamp stand. The diameter of the Perspex panel was 20 mm, meaning that the whole eye was illuminated. The diffusion filter served to make the stimuli appear more uniform (Lisney et al., 2011). Combinations of 25% and 50% ND filters were placed in the light path in order to create ten stimulus light intensity levels across a range of approximately 4 log units (Table 2). The UV-transparency of all of the filters was

confirmed through spectrophotometer measurements (AvaSpec2048 spectrophotometer connected to an Avantes CC-UV/VIS cosine corrector, operated using AvaSoft 7.0 computer software; Avantes Inc., Broomfield, CO). The 10 light intensities were initially quantified in terms of chicken photoreceptor relative quantum catch, using the methods of Rubene et al. (2010) and Lisney et al. (2011). The light spectrum produced by the LEDs was measured on the surface of the Perspex panel using a spectrophotometer. The total amount of light for the part of the spectrum between 300 and 750 nm was then calculated as chicken photoreceptor relative quantum catch for every intensity. Data on relative spectral sensitivity of all five chicken cone types (corrected for filtering effects of oil droplets) and the chicken rod were used in the quantum catch calculations. Relative sensitivity curves were plotted following Govardovskii et al. (2000) and Hart and Vorobyev (2005), using kmax values for the single cones (Hart & Vorobyev, 2005) adult double cones (Hart, Lisney, & Collin, 2006) and rods (Bowmaker et al., 1997). For the light intensity levels at or below the transition from cone to rod vision (which behaviourally occurs at approximately 0.45–1.9 cd m 2 in the chicken; Gover et al., 2009; Lisney et al., 2011), only the rod quantum catch values were used. For the light intensity levels between 1.9 and 10 cd m 2 (corresponding to the mesopic range, estimated using the upper limit for human mesopic vision from Stockman & Sharpe, 2006; also see Lisney et al., 2011) cone and rod quantum catch values were combined. White LEDs alone were then used to create a white, broadband stimulus containing no UV (Lisney et al., 2011; Rubene et al., 2010). The light intensity of this stimulus was varied using ND filters to create ten intensity levels that matched the full spectrum intensity levels in terms of chicken photoreceptor relative quantum catch. A calibrated light meter (Hagner ScreenMaster, B. Hagner AB, Solna, Sweden) was then used to measure the light intensity levels of this white (no UV) stimulus in cd m 2 and lux at the same distance in front of the Perspex panel as the cornea of the eye being stimulated (i.e. 12 mm). This was done to facilitate comparisons between previous studies. The ten light intensities expressed as relative chicken photoreceptor quantum catch, luminance (cd m 2) and illuminance (lx) are given in Table 2. Also, the retinal illumination for each of the ten light intensity levels was estimated in terms of ‘chicken trolands’ (Lisney et al., 2011). 2.4. Stimulus presentation The ten stimuli were divided into two groups based on their intensity levels, ‘group 1’ (stimulus light intensities 1–5, with luminances of 39.7–2740 cd m 2) and ‘group 2’ (stimulus light intensities 6–10 with luminances of 0.7–11.2 cd m 2) (Table 2). The background light levels, measured at the level of the cornea, were

Table 2 Light intensity levels used in this study, expressed as luminance (cd m–2), illuminance (lx), relative cone, rod, and combined photoreceptor quantum catch, and chicken retinal illuminance (chicken trolands). Stimulus light intensity level

Stimulus Group

Luminance (cd m 2)

Illuminance (lx)

Relative cone quantum catch (1023)

Relative rod quantum catch (1022)

Relative photoreceptor quantum catch (1023)

Retinal illuminance (chicken trolands 103)

1 2 3 4 5 6 7 8 9 10

1 1 1 1 1 2 2 2 2 2

2740 1470 558 174 39.7 11.2 5.9 2.5 1.3 0.7

1360 773 351 112 29.0 6.7 3.6 1.5 0.8 0.4

22.2 12.0 4.74 1.52 0.35 0.13 0.07 0.03 – –

– – – – – – 0.18 0.08 0.04 0.02

22.2 12.0 4.74 1.52 0.35 0.13 0.088 0.038 0.004 0.002

166.4 89.3 33.9 10.6 2.51 0.81 0.45 0.20 0.10 0.06

128

T.J. Lisney et al. / Vision Research 62 (2012) 125–133

designed to closely match those used by Lisney et al. (2011), in order not to attenuate contrast in the flickering stimulus. The background light was 3.5 lx during the presentation of the group 1 stimuli and 0.17 lx for the presentation of the group 2 stimuli. Within the two groups of stimuli, the five individual stimulus light intensities were presented randomly. After all of the five stimuli in one group had been presented to the eye, the background illuminance was altered and the eye was allowed to adapt to the new illumination level for approximately 30 min before the next group of five stimuli were presented. Hence, all ten light intensity levels were presented to each eye. The order in which the two groups of stimuli were presented was varied for each hen (Table 1). At any given light intensity, a flickering stimulus was presented for 5–20 s (depending on the number of and length of sweeps used), with an interstimulus interval of 30 s. For each light intensity level, the flicker rate was initially set at approximately 20 Hz less than the FFF established behaviourally by Lisney et al. (2011). The flicker rate was then systematically increased until visual inspection of the ERG waveforms indicated that the retina could no longer produce a modulated electrical signal that remained in phase with the stimulus flicker. Then the same procedure was followed for the next, randomly assigned light level. 2.5. Recording procedure A goldwire electrode (mouse electrode 3 mm, S & V Technologies, Henningsdorf, Germany) in contact with the cornea was used as the active electrode. The reference electrode was a platinum needle electrode (Grass Technologies, West Warwick, RI) inserted subcutaneously approximately 1 cm behind the eye. The electrodes were connected to an ISO-80 bioamplifier (World Precision Instruments, Sarasota, FL) and ERGs were recorded differentially with 10,000 times amplification and the bandpass filter set at 5–1000 Hz. The signal from the amplifier was collected and digitized at a 1000 Hz sampling frequency using a data acquisition system (Powerlab 4/ 30, AD Instruments, Colorado Springs, CO). ERG recordings and stimulus presentations were viewed and controlled using Scope v4.1.3 software (also AD Instruments). Data were averaged across 20–40 sweeps and were smoothed using a modified moving average procedure that involved averaging each sample point with 30 equally weighted points to each side of it (Anon, 2008). The smoothing procedure was performed using the ‘Computed functions’ feature in the Scope software. 2.6. Methods for determining FFF Two methods were used to determine FFF. For the first method (‘method 1’) the ERG waveforms were visual inspected in order to assess whether they remained in phase with the flickering stimuli. For any given light intensity level, the FFF was defined as the frequency in Hz at which the eye was deemed to no longer produce a modulated electrical signal that remained in phase with the stimulus flicker (e.g. Frank, 1999; Hamasaki, 1967) (Fig. 2). The second method (‘method 2’) was adapted from that used by Rubin and Kraft (2007) (Fig. 3). For every light intensity level, average peakto-trough amplitude values (lV) calculated for each stimulus flicker frequency were plotted against stimulus flicker frequency (Hz). A least squares linear regression line was then fitted to the data. For all flicker frequencies at all stimulus intensities there was a negative relationship, i.e. as flicker frequency increased, ERG amplitude decreased. The average background noise amplitude was also determined during each experiment and this allowed a criterion to be set for each experiment. The criterion used was the average noise plus one standard deviation. The criterion was different for every experiment because of differences in background noise between experiments and varied from 3.2 to

6.7 lV. For this second method, the FFF was defined as the frequency at which the linear regression line intersected the criterion line, plus one Hz. For both methods, the highest FFF recorded during an experiment, irrespective of light intensity, was termed the CFF. 2.7. Data analysis The FFF values generated using the two methods for each genotypes of hen were plotted against light intensity (I). The resultant I/FFF curves were doubled-branched, featuring a clearly identifiable break point characteristic of the transition from rod to cone vision (Branchek, 1984; Dodt & Wirth, 1953; Hecht & Shlaer, 1936; Lisney et al., 2011). Graphpad Prism 4.00 software (GraphPad Software, San Diego, CA) was used to fit second-order polynomial functions to the rod and cone branches of each I/FFF curve independently (Lisney et al., 2011). Differences in these functions were tested for statistically with F-tests, both between methods and between genotypes, using the same GraphPad software. 3. Results 3.1. General description Examples of ERG waveforms recorded in response to different light intensity and flicker rates are shown in Fig. 2. FFF increased with increasing light intensity. Across the two methods, FFF values ranged from 12 Hz in hen LSL2, in response to the lowest light intensity (0.7 cd m 2) to a peak (i.e. the CFF) of 119 Hz in hens LB3 and LSL4, in response the highest light intensity (2740 cd m 2). Overall, the average I/FFF curves generated using the two different methods were very similar in shape for both the LB and the LSL hens (Fig. 4). A break point was easily identified in all of the I/FFF curves and this occurred at between 2.5 and 5.9 cd m2. In each I/FFF curve, the rod branch (to the left of the break point) was relatively flat, with FFF values ranging from 14 to 38 Hz. The cone branch, situated to the right of the break point, contained average FFF values that ranged from 36 Hz to a CFF of 107 Hz. 3.2. Comparison of methods For both genotypes of hen, the FFF values obtained using method 2 were slightly higher across the majority of light intensities (4 and 5 Hz higher, on average, for the LB and LSL hens, respectively) (Fig. 4). At the highest light intensity, however, the CFF values were very similar and only differed by 1–2 Hz between the two methods. Second-order polynomial functions were fitted to the rod and cone branches of the I/FFF curves (Fig. 4). These functions were not significantly different between methods for both genotypes of hen: LB hens; rod branch; F3,2 = 14.70; P = 0.0644; cone branch; F3,8 = 3.130; P = 0.0875; LSL hens; rod branch; F3,2 = 5.476; P = 0.1583; cone branch; F3,8 = 3.930; P = 0.0540. 3.3. Comparison between genotypes and individuals As stated above, the average I/FFF curves for the LB and the LSL hens were very similar (Fig. 4). There was no significant difference in the second-order polynomial functions fitted to the rod and cone branches of the I/FFF curves generated using either method 1 (rod branch; F3,2 = 4.159; P = 0.1999; cone branch; F3,8 = 3.548; P = 0.0675) or method 2 (rod branch; F3,2 = 1.077; P = 0.5146; cone branch; F3,8 = 3.130; P = 0.0875) between genotypes. Irrespective of the method used to determine FFF and CFF, a fairly high degree of individual variation in FFF and CFF values

129

T.J. Lisney et al. / Vision Research 62 (2012) 125–133

Fig. 2. Representative examples of electroretinograms (ERGs) recorded from hen LB3 in response to stimuli (S) of different intensities and flicker rates. These examples illustrate the first method (‘method 1’) used to determine FFF/CFF in this study. (a–d) ERG responses to a 174 cd m 2 stimulus at higher flicker frequencies (93–96 Hz). In (a) (93 Hz) and (b) (94 Hz) the ERG response follows the stimulus flicker in phase, but in (c) (95 Hz) and (d) (96 Hz) it does not, resulting in ‘missing’ cycles (indicated by asterisks). In this case, the FFF was 95 Hz. Similar results are shown in the ERG responses to a 2.5 cd m 2 stimulus at lower flicker frequencies (25–29 Hz). In (e) (25 Hz) and (f) (26 Hz) the ERG response follows the stimulus flicker in phase, but in (g) (27 Hz) and (h) (29 Hz) ‘missing’ cycles (marked by asterisks) are again evident, meaning the FFF was 27 Hz. Note that the ERGs shown in (a–h) were averaged across 40 sweeps and smoothed using a modified moving average procedure as described in the text.

(a)

(b)

peak

criterion

trough

Fig. 3. Diagrams illustrating the second method (‘method 2’) used to determine FFF/CFF in this study. For every light intensity level, peak-to-trough amplitudes (a) were measured from ten ERG waveforms for each stimulus flicker frequency. Average peak-to-trough ERG amplitude (±1 standard deviation) was plotted against stimulus flicker frequency and a least squares linear regression line was then fitted to the data (b). The average background noise amplitude was also determined and this allowed a criterion (the average noise amplitude plus 1 standard deviation) to be set (b). The FFF was defined as the frequency at which the linear regression line intersected the criterion line, plus one Hz. The example shown in (b) is for hen LB3 in response to a 174 cd m 2 stimulus (as in Fig. 2a–d). The least squares regression equation is y = 0.7858x + 83.503 (R2 = 0.984) and the criterion value is 6.3 lV. The linear regression line intersects the criterion line at 98 Hz (arrowed), meaning the FFF was 99 Hz.

for both genotypes of hen at stimuli of 5.9 cd m 2 or greater was evident, as indicated by the error bars in Fig. 4. For example, the FFF values determined using the first method for a 558 cd m 2 stimulus varied by 40 Hz, from 71 Hz in hen LB1 to 111 Hz in

hen LB3. Similarly, CFF values determined using the second method and recorded in response to the highest light intensity (2740 cd m 2) ranged from 86 Hz in hen LSL2 to 118 Hz in hen LSL4, representing a 32 Hz difference.

130

T.J. Lisney et al. / Vision Research 62 (2012) 125–133

4.2. No evidence of inter-genotype differences

(a) 140 FFF/CFF (Hz)

120

Cone branch

Rod branch

100 80 60 40 20 0 -0.5

0

0.5

1

1.5

2

2.5

3

3.5

4

3

3.5

4

log10 luminance (cd m-2)

(b) 140 FFF/CFF (Hz)

120

Cone branch

Rod branch

100 80 60 40

Lisney et al. (2011) speculated that differences in behaviourallyderived flicker sensitivity values between studies may be in part due to differences in temporal resolution between different genotypes (breeds) of chicken. Such differences may arise as a consequence of the artificial selection process if visual system traits are genetically or functionally linked to the selected traits. In fish and mammals, visual deficits have been associated with mutation in the PMEL17 gene (Karlsson et al., 2009), which is responsible for the white plumage colour in White Leghorn chickens (Kerje et al., 2004). We found no evidence of differences in the I/FFF curves between the LSL and LB genotypes, which is consistent with other studies that have compared aspects of visual function in different chicken genotypes and found no evidence of genotype-specific visual impairment (DeMello, Foster, & Temple, 1992; Karlsson et al., 2009; Schmid & Wildsoet, 1998). Importantly, however, these studies all used behavioural techniques to test for inter-genotype differences in visual function. Chickens are known to display inter-genotype differences in behavioural characteristics such as fearfulness, motivation and cognitive ability (Jones, 1996; Lindqvist & Jensen, 2009), which could confound using behavioural tests (Lisney et al., 2011). Hence, a controlled experiment where chickens of different genotypes were housed and raised under the same conditions and then tested using ERGs may be the only way to ascertain whether differences in temporal resolution exist among genotypes.

20 0 -0.5

0

0.5

1

1.5

2

2.5

log10 luminance (cd m-2) Fig. 4. Average I/FFF curves for Lohmann Selected Leghorn (LSL) (a) and Lohmann Brown (LB) (b) hens. The data points represented by grey circles and downward facing error bars are FFF/CFF values determined using method 1, while the data points represented by black circles and upward facing error bars are FFF/CFF values determined using method 2. The error bars represent ±1 standard deviation. The two branches (rod and cone) of each of the I/FFF curves have been fitted with secondorder polynomial functions as follows: (a) method 1 data (dashed lines); rod branch, y = 23.305x2 + 28.782x + 18.485, R2 = 0.97; cone branch, y = 2.4286x2 + 40.336x 2.9053, R2 = 0.99; method 2 data (solid lines); rod branch, y = 17.549x2 + 17.605x + 27, R2 = 0.84; cone branch, y = 8.0749x2 + 61.586x 12.644, R2 = 0.991; (b) method 1 data (dashed lines); rod branch, y = 7.8539x2 + 16.984x + 23.011, R2 = 0.98; cone branch, y = 5.5653x2 + 48.128x + 1.2465, R2 = 0.99; method 2 data (solid lines); rod branch, y = 6.0575x2 + 18.396x + 27.259, R2 = 0.99; cone branch, y = 7.985x2 + 56.085x + 2.2266, R2 = 0.99.

4.3. Comparisons with behavioural studies Fig. 5 shows the average, combined I/FFF curve for both hen genotypes obtained using method 2, alongside I/FFF curves obtained from previous behavioural experiments with chickens (Table 3), which have also used LEDs to create full spectrum (i.e. white light with addition of UV) stimuli like those used in this study. The CFF obtained from ERG recordings is approximately 20 Hz higher than the highest average CFF value from a behavioural study (87 Hz; Lisney et al., 2011), and for light intensities

4. Discussion 4.1. The I/FFF curve The ERG-derived chicken I/FFF curve is double-branched, reflecting the presence and activity of both rods and cones (Dodt & Wirth, 1953; Hecht & Shlaer, 1936) and confirming the findings of a previous behavioural study (Lisney et al., 2011). The average CFFs reported here range between 102 and 106 Hz and were recorded in response to the highest stimulus light intensity level used in this study (2740 cd m 2). When considering the hens individually, CFFs of 118–119 Hz were recorded from three individual hens to the same stimulus light intensity level. The cone branches of the I/FFF curves show signs of plateauing at the highest light intensities (Fig. 4), suggesting that the CFF values we report here are close to the true CFF for these particular hens under these specific experimental conditions. In a previous behavioural study, the average CFF measured (87 Hz) was found in response to a 1375 cd m 2 stimulus, followed by a slight decrease in FFF (83 Hz) in response to a 2812 cd m 2 (Lisney et al., 2011).

Fig. 5. Comparison of the average, combined I/FFF curve for both hen genotypes obtained using method 2 from this study with I/FFF curves from previous behavioural studies on chicken that have used similar, LED-generated full spectrum (i.e. white light plus UV) stimuli. The I/FFF curves are presented as polygons that encompass the average values ±1 standard deviation. Key to polygon shading: Black: this study; mid grey: Lisney et al., 2011; light grey: Rubene et al., 2010; full spectrum: dark grey: Rubene et al., 2010; white light. Additional information is provided in Table 3. The dashed line represents the luminance (250 cd m 2) at which the FFFs for each study are compared in Table 3.

131

T.J. Lisney et al. / Vision Research 62 (2012) 125–133

Table 3 Information on the genotypes and ages of chickens and stimulus type (full spectrum [i.e. white light plus UV] or white light alone) used in three studies of FFF/CFF in chickens, the results of which are compared in Fig. 5. In all three studies, LEDs were used to create the stimuli. Study

This study

Lisney et al. (2011)

Rubene et al. (2010)

Rubene et al. (2010)

Genotype

Lohmann Brown (LB) and Lohmann Selected Leghorn (LSL) 69–73 weeks Full spectrum 87.0

Old Swedish game breed ‘Gammalsvensk dvärghöna’ 2–4 years Full spectrum 74.6

White leghorn (Bovans) 5–12 weeks Full spectrum 62.3

White leghorn (Bovans) 5–12 weeks White 55.0

Black

Mid grey

Light grey

Dark grey

Age Stimulus type FFF at stimulus luminance of 250 cd m 2 (Hz) Shading in Fig. 5

>250 cd m 2, the ERG FFF values are all higher than those from the behavioural studies. Furthermore, the range over which the transition from rod to cone vision occurs as found in this study is also slightly higher than the range reported previously in behavioural studies (Gover et al., 2009; Lisney et al., 2011). Because FFF/CFF can be influenced by a number of factors relating to the stimulus (e.g. size, intensity, colour spectrum) as well the physiological condition of the experimental subjects (Landis, 1954), caution should be applied when comparing the results of different studies. Nevertheless, the stimuli used in this study were similar to those used in previous behavioural studies (Lisney et al., 2011; Rubene et al., 2010) and in all studies healthy individuals were used for the experiments. Therefore we are confident that the behaviourallyand ERG-derived measures of the FFF, CFF and rod-cone transition in chicken reflect differences in the conscious perception of flicker vs. the physiological responses of the retina. Behavioural studies (Lisney et al., 2011; Rubene et al., 2010) have shown that hens cannot consciously perceive flicker above ca. 100 Hz in full spectrum light, whereas the results from this study show that the retina at least can respond to flicker frequencies in the 100–120 Hz range. Similar results have been reported in the pigeon, C. livia; Hendricks (1966) reported a behaviourally-derived CFF of 77 Hz in this species, whereas using ERG, Dodt and Wirth (1953) found a CFF of 143 Hz. These results suggest that, in birds, temporal resolution is not limited by the retina’s ability to resolve flickering stimuli, but rather that temporal summation occurs further along the visual pathway in the brain. In mammals, the retina is able to follow stimuli flickering at higher frequencies than the visual cortex (Eysel & Burandt, 1984; Lindsley, 1953; van de Grind, Grüsser, & Lunkenheimer, 1973; Walker et al., 1943), and in turn, the visual cortex is able to follow flickering stimuli at higher frequencies than the behaviourally-derived CFF (Schneider, 1968; Schwartz & Lindsley, 1964). There is evidence, however that at least some high flicker frequency information does reach the visual cortex without necessarily resulting in flicker perception (Eysel & Burandt, 1984; van der Tweel & Verduyn Lunel, 1965; Williams et al., 2004). According to Schneider (1968), FFF/CFF may actually be a function of an animal’s inability to use all of the stimulusrelated information that can be detected by the visual system, but the underlying reasons for differences in the perceptual vs. physiological detection of flickering stimuli are not well understood. Compared to mammals, even less is known about the processing of flicker in the avian brain. Therefore, comparisons of FFF/CFF values obtained using behavioural tests, ERGs and electrophysiological recordings from visual brain areas in birds, such as the optic tectum and the visual Wulst (which bears a close resemblance and may be homologous to the mammalian primary visual cortex; Karten et al., 1973; Medina & Reiner, 2000; Shien Wei Ng et al., 2010) will be insightful and may prove important in understanding the processing of temporal visual information in both mammalian and avian brains.

4.4. Welfare implications of retinal responses to >100 Hz Although hens do not appear to be able to consciously perceive flicker above approximately 90 Hz (Lisney et al., 2011), the retina’s ability to respond to flicker at frequencies >100 Hz may still result in distress for the animals. In humans, exposure to such so-called ‘‘invisible flicker’’ can cause headaches, eye-strain, anxiety and changes in eye-saccades (Wilkins, Veitch, & Lehman, 2010) and may affect the brain (Küller & Laike, 1998) and the immune system (Martin, 1989). Therefore the potential effects of flicker >100 Hz on poultry welfare should not be underestimated until more information is available. Having said this, the lighting used in industrial poultry farming differs from the full spectrum stimulus used in this study in some important ways. Firstly, fluorescent lamps flicker with a considerably shallower modulation depth than the 100% modulation used in the present study. The ability to resolve flicker increases with the modulation depth of the light source (Jarvis et al., 2002; Nuboer, Coemans, & Vos, 1992) and so it remains to be seen whether the chicken ERG can respond to 100 Hz flicker at such reduced modulation. Secondly, the duty cycle of fluorescent lamps is usually higher than the 50% used in this study. As a low lighting duty cycle may in humans produce discomfort compared to a high duty cycle (Bullough et al., 2011) one might speculate that fluorescent flicker is less likely than our stimulus to be resolved by the chicken outer retina, although it is currently unknown what effect, if any, variation in the duty cycle has on the shape of avian I/FFF curves. However, for reasons of energy efficiency, magnetic ballasts (which produce 100 or 120 Hz flicker) to control fluorescent lamps are being phased out in Europe (The Commission of the European Communities, 2009a, 2009b) and the USA (US Congress & Natural Resources, 2005). LED lighting is a promising energy-efficient alternative (Pimputkar, Speck, DenBaars, & Nakamura, 2009). Various existing techniques to control LED lighting can produce flicker with modulation depths and duty cycles that are closer to those of our stimulus light (Wilkins, Veitch, & Lehman, 2010). Thirdly, we used a full spectrum stimulus (i.e. white light with addition of UV) that was designed to be comparable with the International Commission on Illumination (CIE) standard data on the spectral power distribution of sunlight (D65) (Rubene et al., 2010). In contrast, fluorescent lamps emit very little or no UV (Lewis & Morris, 2006). White stimuli with UV have been shown to improve flicker detection rates in hens compared to similar stimuli without UV (Rubene et al., 2010) (also see Fig. 5). Under normal fluorescent lighting therefore, we predict that the chicken ERG should show lower FFFs and hence be less likely respond to 100 or 120 Hz flicker, than indicated by the results of the present study. Again this is yet to be ascertained but should be a goal of future research. There is currently little direct evidence that fluorescent lighting, or the use of LEDs, is particularly detrimental to the welfare of poultry. Indeed, hens even seem to prefer fluorescent lighting to incandescent (Sherwin, 1999; Widowski, Keeling, & Duncan,

132

T.J. Lisney et al. / Vision Research 62 (2012) 125–133

1992). However, in more than one study, the potential effects of lamp flicker have been confounded by the failure to adequately control for variation in the spectral composition and/or intensity of the light (reviewed by Greenwood et al. (2004), Lewis and Morris (2006) and Smith (2003). Because, incandescent lighting is being phased out across the world (The Commission of the European Communities, 2009b), the use of fluorescent lamps, and possibly LEDs, is likely to further increase, there is a need for more and better-controlled studies on the effects of artificial lighting on the behaviour and physiology of poultry. Acknowledgments Constructive comments from two anonymous reviewers greatly helped to improved the manuscript. We are grateful to Heidi Lindfeldt and Jenny Nielsen for animal husbandry. This study was mainly financed by grants from the Swedish Research Council Formas to AÖ. T.J.L. was supported by a post-doctoral fellowship from the Carl Tryggers Foundation for Scientific Research and acknowledges the support of Dr. Douglas Wylie (University of Alberta) during the analysis and write-up of this research. References Anon (2008). Scope v4.1 user’s guide. Bella Vista: ADInstruments Pty Ltd.. Bowmaker, J. K., Heath, L. A., Wilkie, S. E., & Hunt, D. M. (1997). Visual pigments and oil droplets from six classes of photoreceptor in the retinas of birds. Vision Research, 37, 2183–2194. Branchek, T. (1984). The development of photoreceptors in the zebrafish, Brachydanio rerio. II. Function. Journal of Comparative Neurology, 224, 116–122. Brown, J. L. (1965). Flicker and intermittent stimulation. In C. H. Graham (Ed.), Vision and visual perception (pp. 251–320). New York: John Wiley & Sons. Bullough, J. D., Sweater Hickcox, K., Klein, T. R., & Narendran, N. (2011). Effects of flicker characteristics from solid-state lighting on detection, acceptability and comfort. Lighting Research & Technology, 43, 337–348. D’Eath, R. B. (1998). Can video images imitate real stimuli in animal behaviour experiments? Biological Reviews, 73, 267–292. DeMello, L. R., Foster, T. M., & Temple, W. (1992). Discriminative performance of the domestic hen in a visual acuity task. Journal of the Experimental Analysis of Behavior, 58, 147–157. Dodt, E., & Enroth, C. (1954). Retinal flicker response in cat. Acta Physiologica Scandinavica, 30, 375–390. Dodt, E., & Wirth, A. (1953). Differentiation between rods and cones by flicker electroretinography in pigeon and guinea pig. Acta Physiologica Scandinavica, 30, 80–89. Douglas, R. H., & Hawryshyn, C. W. (1990). Behavioural studies of fish vision: An analysis of visual capabilities. In R. H. Douglas & M. B. A. Djamgoz (Eds.), The visual system of fish (pp. 373–418). London: Chapman and Hall. Eysel, U. T., & Burandt, U. (1984). Fluorescent tube light evokes flicker responses in visual neurons. Vision Research, 24, 943–948. Frank, T. M. (1999). Comparative study of temporal resolution in the visual systems of mesopelagic crustaceans. Biological Bulletin, 196, 137–144. Govardovskii, V. I., Fyhrquist, N., Reuter, T., Kuzmin, D. G., & Donner, K. (2000). In search of the visual pigment template. Visual Neuroscience, 17, 509–528. Gover, N., Jarvis, J. R., Abeyesinghe, S. M., & Wathes, C. M. (2009). Stimulus luminance and the spatial acuity of domestic fowl (Gallus g. domesticus). Vision Research, 49, 2747–2753. Greenwood, V. J., Smith, E. L., Goldsmith, A. R., Cuthill, I. C., Crisp, L. H., Walter-Swan, M. B., et al. (2004). Does the flicker frequency of fluorescent lighting affect the welfare of captive European starlings? Applied Animal Behaviour Science, 86, 145–159. Hamasaki, D. I. (1967). An anatomical and electrophysiological study of the retina of the owl monkey, Aotes trivirgatus. Journal of Comparative Neurology, 130, 163–174. Hart, N. S., Lisney, T. J., & Collin, S. P. (2006). Cone photoreceptor oil droplet pigmentation is affected by ambient light intensity. Journal of Experimental Biology, 209, 4776–4787. Hart, N. S., & Vorobyev, M. (2005). Modelling oil droplet absorption spectra and spectral sensitivities of bird cone photoreceptors. Journal of Comparative Physiology A, 191, 381–392. Hecht, S., & Shlaer, S. (1936). Intermittent stimulation by light. V. The relation between intensity and critical frequency for different parts of the spectrum. Journal of General Physiology, 19, 965–977. Hendricks, J. (1966). Flicker thresholds as determined by a modified condition suppression procedure. Journal of the Experimental Analysis of Behavior, 9, 501–506. Henkes, H. E. (1964). Recent advances in flicker-electroretinography. Documenta Ophthalmologica, 18, 307–314.

Jarvis, J. R., Taylor, N. R., Prescott, N. B., Meeks, I., & Wathes, C. M. (2002). Measuring and modelling the photopic flicker sensitivity of the chicken (Gallus g. domesticus). Vision Research, 42, 99–106. Jones, R. B. (1996). Fear and adaptability in poultry: Insights, implications and imperatives. World’s Poultry Science Journal, 52, 131–174. Jones, M. P., Pierce, K. E., & Ward, D. (2007). Avian vision: A review of form and function with special consideration to birds of prey. Journal of Exotic Pet Medicine, 16, 69–87. Karlsson, A. C., Kerje, S., Halböök, F., & Jensen, P. (2009). The dominant white mutation in the PMEL17 gene does not cause visual impairment in chickens. Veterinary Ophthalmology, 12, 292–298. Karten, H. J., Hodos, W., Nauta, W. J. H., & Revzin, A. M. (1973). Neural connections of the ’’visual Wulst’’ of the avian telencephalon. Experimental studies in the pigeon (Columbia livia) and owl (Speotyto cunicularia). Journal of Comparative Neurology, 150, 253–277. Kerje, S., Sharma, P., Gunnarsson, U., Kim, H., Bagchi, S., Fredriksson, R., et al. (2004). The Dominant white, Dun and Smoky color variants in chicken are associated with polymorphisms in the PMEL17 gene. Genetics, 168, 1507–1518. Kjaer, J. B., & Vestergaard, K. S. (1999). Development of feather pecking in relation to light intensity. Applied Animal Behaviour Science, 62, 243–254. Küller, R., & Laike, T. (1998). Impact of flicker from fluorescent lighting on wellbeing, performance, and physiological arousal. Ergonomics, 41, 433–447. Landis, C. (1954). Determinants of the critical flicker fusion threshold. Physiological Reviews, 34, 259–286. Lewis, P. D., Morris, T. R. (2006). Poultry lighting: The theory and practice. Northcot, Andover. Lindqvist, C., & Jensen, P. (2009). Domestication and stress effects on contrafreeloading and spatial learning performance in red jungle fowl (Gallus gallus) and White Leghorn layers. Behavioural Processes, 81, 80–84. Lindsley, D. B. (1953). Effect of photic stimulation in visual pathways from retina to cortex. Science, 117, 469. Lisney, T. J., Rubene, D., Rózsa, J., Løvlie, H., Håstad, O., & Ödeen, A. (2011). Behavioural assessment of flicker fusion frequency in chicken Gallus gallus domesticus. Vision Research, 51, 1324–1332. Lu, J., Zoran, M. J., & Cassone, V. M. (1995). Daily and circadian variation in the electroretinogram of the domestic fowl: Effects of melatonin. Journal of Comparative Physiology A, 177, 299–306. Maddocks, S. A., Goldsmith, A. R., & Cuthill, I. C. (2001). The influence of flicker rate on plasma corticosterone levels in European starlings, Sturnus vulgaris. General and Comparative Endocrinology, 1245, 315–320. Martin, P. (1989). Psychoimmunology: Relations between brain, behavior, and immune function. In P. P. G. Bateson & P. H. Klopfer (Eds.). Perspectives in ethology (Vol. 8, pp. 173–214). New York: Plenum. Medina, L., & Reiner, A. (2000). Do birds possess homologues of mammalian primary visual, somatosensory and motor cortices? Trends in Neurosciences, 23, 1–12. Meyer, D. B. (1977). The avian eye and its adaptations. In F. Crescitelli (Ed.), Handbook of sensory physiology VII/5 (pp. 549–611). Berlin: Springer-Verlag. Nuboer, J. F. W., Coemans, M. A. J. M., & Vos, J. J. (1992). Artificial lighting in poultry houses: Do hens perceive the modulation of fluorescent lamps as flicker? British Poultry Science, 33, 123–133. Ordy, J. M., & Samorajski, T. (1968). Visual acuity and ERG-CFF in relation to the morphologic organization of the retina among diurnal and nocturnal primates. Vision Research, 8, 1205–1225. Pimputkar, S., Speck, J. S., DenBaars, S. P., & Nakamura, S. (2009). Prospects for LED lighting. Nature Photonics, 3, 180–182. Prescott, N. B., Wathes, C. M., & Jarvis, J. R. (2003). Light, vision and the welfare of poultry. Animal Welfare, 12, 269–288. Railton, R. C. R., Foster, T. M., & Temple, W. (2009). A comparison of two methods for assessing critical flicker fusion frequency in hens. Behavioural Processes, 80, 196–200. Rubene, D., Håstad, O., Tauson, R., Wall, H., & Ödeen, A. (2010). Presence of UV wavelengths improves temporal resolution of the avian visual system. Journal of Experimental Biology, 213, 3357–3363. Rubin, G. R., & Kraft, T. W. (2007). Flicker assessment of rod and cone function in a model of retinal degeneration. Documenta Ophthalmologica, 115, 165–172. Schmid, K. L., & Wildsoet, C. F. (1998). Assessment of visual acuity and contrast sensitivity in the chick using an optokinetic nystagmus paradigm. Vision Research, 38, 2629–2634. Schneider, C. W. (1968). Electrophysiological analysis of the mechanisms underlying critical flicker frequency. Vision Research, 8, 1235–1244. Schwartz, A. S., & Lindsley, D. B. (1964). Critical flicker fusion frequency and photic following in the cat. Boletín del Instituto de Estudios Medicos y Biologicos, Mexico, 22, 249–262. Sherwin, C. M. (1999). Domestic turkeys are not averse to compact fluorescent lighting. Applied Animal Behaviour Science, 64, 47–55. Shien Wei Ng, B., Grabska-Barwinska, A., Güntürkün & Jancke, D. (2010). Dominant vertical orientation processing without clustered maps: Early visual brain dynamics imaged with voltage-sensitive dye in the pigeon visual Wulst. Journal of Neuroscience, 30, 6713–6725. Smith, C. P. (1992). Animal models in biomedical research: Poultry. Beltsville: United States Department of Agriculture. Smith, E. L. (2003). Effect of the visual environment on avian welfare. Unpublished PhD Thesis. University of Bristol: Bristol. Stockman, A., & Sharpe, L. T. (2006). Into the twilight zone: The complexities of mesopic vision and luminous efficiency. Ophthalmic Physiological Optics, 26, 225–239.

T.J. Lisney et al. / Vision Research 62 (2012) 125–133 Tauson, R., & Holm, K. -E. (2001). First furnished small group cages for laying hens in evaluation program on commercial farms in Sweden. In H. Oester, C. Wyss (Eds.), Proceedings of the 6th European symposium on poultry welfare, Zollikofen, Switzerland (pp. 26–32). Swiss Branch of the WPSA, Zollikofen, Switzerland. The Commission of the European Communities, 2009a. COMMISSION REGULATION (EC) No 245/2009 of 18 March 2009 implementing Directive 2005/32/EC of the European Parliament and of the Council with regard to ecodesign requirements for fluorescent lamps without integrated ballast, for high intensity discharge lamps, and for ballasts and luminaries able to operate such lamps, and repealing Directive 2000/55/EC of the European Parliament and of the Council. Official Journal of the European Union L 76/17. The Commission of the European Communities, 2009b. COMMISSION REGULATION (EC) No. 244/2009 of 18 March 2009 implementing Directive 2005/32/EC of the European Parliament and of the Council with regard to ecodesign requirements for non-directional household lamps. Official Journal of the European Union L 76/3. US Congress, Committee on energy and natural resources, 2005. Energy Policy Act of 2005. 109th Cong., 1st sess. 24 January 2012. . van de Grind, W. A., Grüsser, O.-J., & Lunkenheimer, H.-U. (1973). Temporal transfer properties of the afferent visual system. Psychological, neurophysiological and theoretical investigations. In R. Jung (Ed.). Handbook of sensory physiology (Vol. VII/3, pp. 432–573). Berlin: Springer-Verlag.

133

van der Tweel, L. H., & Verduyn Lunel, H. F. E. (1965). Human visual responses to sinusoidally modulated light. Electroencephalography and Clinical Neurophysiology, 18, 587–598. Walker, A. E., Woolf, J. I., Halstead, W. C., & Case, T. J. (1943). Mechanisms of temporal fusion effect of photic stimulation on electrical activity of visual structures. Journal of Neurophysiology, 6, 13–219. Widowski, T. M., Keeling, L. J., & Duncan, I. J. H. (1992). The preferences of hens for compact fluorescent over incandescent lighting. Canadian Journal of Animal Science, 72, 203–211. Wilkins, A. J., & Clark, C. (1990). Modulation of light from fluorescent lamps. Lighting Research & Technology, 22, 103–109. Wilkins, A., Veitch, J., & Lehman, B. (2010). LED lighting flicker and potential health concerns: IEEE standard PAR1789 update. In Proceedings of the IEEE energy conversion congress and exposition (ECCE 2010), Atlanta, USA, September 12–16, 2010, pp. 171–178. Williams, P. E., Mechler, F., Gordon, J., Shapley, R., & Hawken, M. J. (2004). Entrainment to video displays in primary visual cortex of macaque and humans. Journal of Neuroscience, 24, 8278–8288. Wortel, J. F., Rugenbrink, H., & Nuboer, J. F. W. (1987). The photopic spectral sensitivity of the dorsal and ventral retinae of the chicken. Journal of Comparative Physiology A, 160, 151–154.