Myopia, most often caused by elongation of the eye, is a leading, and increasing, cause of vision impairment globally. Existing evidence indicates retinal signalling has a key role in driving ocular growth. Genome-wide association studies have identified many genetic polymorphisms associated with myopia in the general population. Several of these are near genes expressed in the retina and involved in retinal physiology. The common variant most strongly associated with myopia is at locus rs524952 near the GJD2 gene, which encodes the Connexin-36 protein, forming retinal gap junctions. The electroretinogram (ERG) represents the electrical response of the retina to light stimuli, and it can be recorded non-invasively from the living human eye. Here, we present findings from our studies investigating associations between myopia risk polymorphisms and ERG parameters. TwinsUK comprises several thousand adult twins who have volunteered to participate in research studies based at St Thomas’ Hospital in London. Initially, dark-adapted and light-adapted ERGs were recorded in response to international standard and experimental stimulus protocols in over 200 twins, following pharmacological pupil dilation (1.0% tropicamide supplemented in most cases with 2.5% phenylephrine), using a conductive fibre electrode placed in the lower conjunctival fornix. Stimuli were delivered using the Colordome (Diagnosys UK, Cambridge, UK). In genotyped individuals (n=186), we specifically investigated associations between allelic dosage at rs524952 and ERG parameters (amplitudes and peak times of a-waves, b-waves and 30 Hz flicker responses), using a mixed linear model, adjusting for age, sex and familial relatedness. We found significant associations with parameters relating to cone-driven retinal signals. Taken together with findings in patients with selective loss of post-receptoral signals, we found evidence of a specific association with cone-driven OFF bipolar cell signals. Subsequently, we analysed ERG recordings from over 1000 twins, made with a portable device (RETeval system, LKC technologies, Gaithersburg, MD) and using skin electrodes. These were responses to 30 Hz flickering stimuli delivered through natural pupils, but with stimulus and background strength adjusted according to pupil diameter to deliver retinal illuminance equivalent to international standards. In genotyped individuals (n=895), we explored associations between 334 known myopia-risk loci and ERG flicker peak times. Although no association achieved statistical significance after correction for multiple testing, one of the top loci attaining nominal significance was rs13268738, within the CNGB3 gene (which encodes a subunit of the cyclic nucleotide-gated channel in the outer segments of cone photoreceptors). This specific locus was then examined in the group of participants who had undergone mydriatic recordings with the conductive fibre electrode: allelic dosage was found to be significantly associated with flicker peak times in this groups also, thus replicating the association found in the larger group. Overall, our findings highlight possible pathways through which these particular myopia risk loci might be acting, supporting a role for alterations in retinal cone-driven signalling in myopia development.