The hypothesis that the risk of stochastic effects is directly proportional to the dose for all levels of dose and dose rate (below those at which deterministic effects occur). i.e. that any non-zero dose implies a non-zero risk of stochastic effects. This is the working hypothesis on which the Agency’s safety standards (and ICRP’s recommendations) are based. It is not proven - indeed it is probably not provable - for low doses and dose rates, but it is considered the most radiobiologically defensible assumption on which to base safety standards. Other hypotheses conjecture that the risk of stochastic effects at low doses and/or dose rates is: (a) greater than that implied by the LNT hypothesis (superlinear hypotheses); (b) less than that implied by the LNT hypothesis (sublinear hypotheses); (c) zero below some threshold value of dose or dose rate (threshold hypotheses); or (d) negative below some threshold value of dose or dose rate, i.e. that low doses and dose rates protect individuals against stochastic effects and/or other types of harm (hormesis hypotheses)